Acid Rain, by David Newbery

Centre for Economic Policy Research
Center for Economic Studies

Maison des Sciences de l'Homme
Acid Rain
Author(s): David M. Newbery, Horst Siebert and John Vickers
Source: Economic Policy, Vol. 5, No. 11 (Oct., 1990), pp. 297-346
Published by: Wiley on behalf of the Centre for Economic Policy Research, Center for Economic
Studies, and the Maison des Sciences de l'Homme
Stable URL: http://www.jstor.org/stable/1344480 .
Accessed: 08/09/2014 12:37
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
.
Wiley, Centre for Economic Policy Research, Center for Economic Studies, Maison des Sciences de l'Homme
are collaborating with JSTOR to digitize, preserve and extend access to Economic Policy.
http://www.jstor.org
This content downloaded from 193.54.67.93 on Mon, 8 Sep 2014 12:37:20 PM
All use subject to JSTOR Terms and Conditions
Economic
Policy
October 1990 Printed in Great Britain
Acid rain
David
Newbery
Summary
Acid rain is not a new
phenomenon,
but environmental awareness
has
grown rapidly
over the last decade. Much data has been collected
and the transmission
process
is better understood.
Policy-makers
in
Europe
have set themselves the
objective of
a
uniform
30% reduction
in national emissions
of sulphur
dioxide and a
freeze
on emissions
of nitrogen
oxides.
Whereas
simple
economic
principles
have
informed
much
of
the
policy
debate in the
US,
the same has not been true in
Europe.
A
reduction
of
emissions which is
uniform
across countries is
likely
to
be
highly inefficient:
rather,
particular
emissions should be curtailed
until the
marginal
cost
of further
abatement
equals
the
marginal
benefit
as measured
by
the
marginal damage thereby
avoided. Within
and across
countries,
marginal
abatement costs
and,
a
fortiori,
marginal damage from
acid rain
vary greatly.
Thus an
efficient
emission-reduction
programme
involves
unequal
reductions
of
emissions across countries and activities. The
paper
offers
calcula-
tions
of
the structure
of
an
efficient programme,
and discusses how
we can best make use
of
markets and
prices
to decentralize as
many
decisions as
possible.
Acid rain
David M.
Newbery
Department
of
Applied
Economics,
Cambridge
and CEPR
1. Introduction
Air
pollution
is not a new
phenomenon-Londoners
in the twelfth
century complained
about the noxious fumes from
burning
sea
coal,
and the corrosive effects of
sulphur
dioxide
(SO2)
dissolved in rain have
been well understood for a least a
century.
But the focus of concern
constantly
shifts. In Britain Smoke Abatement Acts date from 1853-56.
The landmark Clean Air Act of 1956 was
primarily
a
response
to the
health hazards associated with the
unregulated burning
of coal. An
estimated
4,000
people
died in the
great
London
smog
of December
1952. The US has also been concerned with
reducing
coal
pollution,
but was also active in
reducing
automobile
pollution
from
quite early
on. Here the
impetus
was the
deteriorating
air
quality
in urban areas
like Los
Angeles
and
Washington DC,
where
photochemical smog
led
to
high
levels of
ozone,
traced to exhaust emissions of
hydrocarbons
and
nitrogen
oxides
(NOx).
California led the
way
in
introducing tighter
emissions controls. In
Europe,
the
impetus
for environmental
policy
developed
because acid rain - from the
SO2
and
NOx
emissions from
the
burning
of coal and oil
- is no
respecter
of national boundaries.
Each
locality
and
country
discovered that some of the
immediately
harmful effects of
burning
coal could be avoided
by building
tall chim-
neys,
but this
merely dispersed
the
pollutants
elsewhere,
often across
great
distances to other countries. These countries could not
directly
control the
deposition
of acid
rain,
and could instead
only complain
I I
Research
support
from the UK Economic and Social Research Council's
grant
Privatisation and
Reregulation of
the Network Industries is
gratefully acknowledged.
I am indebted to
Margaret
Clark
for assistance with the literature
search,
to Michael Hannaman for
bringing
to
my
notice the
paper by
Maler
(1989),
to R. G. Derwent of Harwell for his
extremely thorough
and
helpful
scientific comments,
and to David
Pearce, John
Vickers and David
Begg
for careful comments.
None of these is to he held
responsible
for the
interpretations
I have chosen.
Acid rain 299
and
negotiate
for some coordinated solution to the
perceived problem.
Different countries
responded
to different facets of the
pollution prob-
lem. The Scandinavian countries were troubled
by
the death and
disap-
pearance
of fish from lakes and rivers. Germans worried about forest
die-back. Glasnost revealed the full extent of the environmental disasters
in Eastern
Europe,
and
provided
the focus for local
hostility
to the
environmental
insensitivity
of central
planning.
Environmental awareness has
grown rapidly
over the
past
decade,
and with it the
growing
realization that we live on a rather small and
fragile planet.
The
green
movement has had to work hard to
capture
the attention and
imagination
of the
public
and
politicians,
and has had
to resort to emotional
arguments
to
get
its
message
across. There is no
doubt that the hidden environmental costs of current
technology may
be
high,
even
life-threatening,
but it is also clear from recent
experience
that the costs of
carelessly designed
environmental
regulation
can also
be
high.
As
economists,
we have a
duty
to
argue
for cost-effective
environmental
policies.
Inefficient
policies
not
only
achieve less than
they
should,
but
they
also run the risk of
alienating taxpayers
and
consumers who
ultimately pay
for the
regulation
and
may
undermine
the aims of the environmental movement.
Economists,
with honourable
exceptions,
have tended to
ignore
environmental economics because it
seems to raise few new ideas. Most of the useful
techniques
have been
the stuff of
undergraduate
welfare economics since
Pigou's day. Though
each
generation
adds to the stock of
knowledge
and
techniques,
the
subject
has not been at the theoretical
cutting edge
for some time. This
might
not have mattered if economists had been
supplied
with accessible
facts with which to clothe the
theory
and to
bring
the
policy
issues into
sharp perspective.
But these facts are
largely produced by
scientists
unfamiliar with the economic
style
of
argument,
and often unconcerned
with economic costs and benefits. There has been too little communica-
tion between the
disciplines.
It is
interesting
to
compare
the situation in the US. The
style
of
regulation exemplified by
the Environmental Protection
Agency,
and
the
separation
of
legislative
and executive
power,
means that environ-
mental
legislation
has to be
argued
in a
quasi-judicial way
before
being
enacted,
and economists have been
centrally
involved in the
ensuing
debates - not
necessarily successfully.
As a result of
having
to make a
quantified
case in
public,
economists have
investigated
the scientific
evidence,
have conducted
empirical enquiries,
and have identified the
gaps
in our
knowledge.
Environmental economics has received a con-
siderable
impetus
and a solid
body
of
theory
and evidence on which to
build. We in
Europe lag
behind, though
there are
signs
that the times
are
changing.
This contribution deals with a small
part
of the environmental
debate,
that concerned with acid rain. It is an
important topic
- not as
important
as the
greenhouse effect,
which is
global
in
scale,
and
probably
not as
important
as traffic
congestion,
which is a domestic matter for each
country. Nevertheless,
substantial sums of
money
have been
spent
and
are now
being
committed in an
attempt
to alleviate the
problem
of acid
rain. The thrust of this
paper
is that this
programme
as
currently
interpreted
is
flawed,
unnecessarily expensive,
and if it
succeeds,
it runs
the risk of
high political
cost.
Relatively simple
economic
principles
applied
to the
appropriate
facts
ought
to be able to achieve the same
environmental benefits at
substantially
lower
cost,
and in a more decen-
tralized and less
politically problematic way.
I make no
apologies
for the
high
ratio of facts to
theory
in what
follows. The environmental debate has been
long
on emotional
argu-
ment and short on substance for too
long.
I am not an
expert
in this
field,
and have had to
rely
on
secondary
sources for the data. On the
crucial issue of
quantifying
the benefits of
reducing
acid rain I have
not been able to find
adequate
evidence and so cannot
finally quantify
the efficient
policy.
But I have found
enough
evidence to cast consider-
able doubt about the
priorities
for
abatement,
and to
suggest
where
research effort should be concentrated. Several
findings surprise
me.
Fish death from acid rain is
sad,
but
economically unimportant.
Tree
death
may
be far more
important, though
there are
worrying
uncer-
tainties about the cause and cure of this
problem.
Health
problems
associated with coal
emissions,
particularly
the combination of
SO2
and
particulates (smoke particles)
are
potentially
of the first
importance,
whereas those associated with
NOx
and ozone seem trivial.
2. Acid rain and its effects
In order to understand the acid rain
problem
it is
necessary
first to
describe the causes and
consequences
of acid rain. Considerable scien-
tific research over the
past
decade has illuminated this
phenomenon,
though
uncertainties remain. The next
step
is to
identify
the sources
and measure the amounts of
pollutant released,
and their destination.
What is it that causes the
damage,
where does the main
damage occur,
and what are
economically
the most
expensive consequences
of acid
rain?
Finally,
one needs to determine the
techniques
available for
reducing emissions,
and the costs of
abatement,
in order to
identify
cost-effective abatement
policies.
This last
step
is
usually ignored by
ecologists
and
politicians,
who are content once
they
have found
ways
of
reducing
acid rain to
press
for the maximum
(politically?)
feasible
degree
of abatement. This section addresses each issue in turn.
300
David
Newbery
2.1.
Defining
acid rain
Acid rain is
normally
understood to include the
deposition
of the acidic
combustion
products sulphur
dioxide, SO2,
various
nitrogen oxides,
NOx,
and
chloride, Cl-,
either as
dry gases
or
particles,
or as wet
deposits
in
rain, snow, sleet, hail,
mist or
fog.
These
pollutants usually undergo
a series of chemical transformations into
sulphuric acid,
nitric acid and
hydrochloric
acid. These acids affect the
environment,
both
directly
and
indirectly
in
causing
the release of further harmful chemicals such
as aluminium. Acid rain can be measured in a
variety
of
ways
- in terms
of tonnes of the
original gases
released,
or tonnes of elemental
sulphur,
or in terms of the
acidity
of the
rainfall, run-off,
streams or lakes.
Acidity
is measured in
pH
units on a
logarithmic
scale.' As the scale is
logarithmic,
rainfall with a
pH
of 5 is 10 times as acid as that with a
pH
of 6.
Unpolluted
rain is
slightly
acidic from dissolved carbon dioxide
and has a
pH
of about 5.6. Sea water is
naturally alkaline, having
a
pH
of 8.3.
Most
SO2
comes from
large
combustion
plants-thus
in
1987,
85%
came from
large
combustion
plants,
and
73%
from
power
stations. Of
UK emissions from fossil fuel
combustion,
79% came from coal combus-
tion and 12%
from fuel oil.2
Sulphur
dioxide
pollutes
the environment
through
two different routes. Much of the
gas
falls to earth within
300 km of the source in its
dry
form,
and this
process
is described as
dry deposition. Long-range transport
occurs because
SO2
is oxidized
to
sulphate particles,
which are not
readily deposited
in
dry
form. Their
main removal is
by scavenging
in
rain-making processes
as wet
deposi-
tion,
which
may
occur
1,000-2,000
km from the source. Wet
deposition
can be
reported
in two
ways
-
by
its
intensity
and cumulative
deposition.
Intensity
is shown
by
the
maps
of the
average acidity
of
precipitation
(in
pH),
and cumulative
deposition by
wet
deposited acidity
in
grams
of
hydrogen
ions
per square
metre
per year. Deposited acidity
is the
product
of the
acidity
of the rainfall and the amount of rain -wetter
areas in the west
may
have more acid
deposited
even
though
the
precipitation
is less acidic.
2.2.
Measuring
acid rain
The
European Monitoring
and Evaluation
Programme (EMEP)
was set
up
in 1978 to monitor the movement of
pollutants,
and to determine
I
I
pH
is defined as the
negative logarithm
of the
hydrogen
ion
(H+) concentration, having
the
perverse
effect that lower numbers
correspond
to
higher acidity.
Thus 0 is most
acid,
7 is
neutral and 14 most alkaline. Lemon
juice
has a
pH
of
2,
milk of
magnesia
10.5.
2
UK data are taken from the
Digest of
Environmental Protection and Water
Statistics, 1988, (Depart-
ment of the
Environment). Figures
after 1970 are based on revised emission factors described
therein.
Acid rain 301
where the
deposition
of
pollutants
released from each source occurs.
Until
recently
the
only pollutant
tracked was
SO2, though
now
NOx
is
also monitored. The surface of
Europe
is divided into
squares
with
grid
lines 150 km
apart.
There are about 720
grid
line intersections on land
and about 100
monitoring
sites which are used in the EMEP model and
these are termed arrival
points. Using
detailed
meteorological
informa-
tion,
the track of air which arrived at each of the 820 or so
points
is
followed backwards in time for 96 hours. An air
parcel
is then studied
forwards in time as it follows each back-track
precisely, picking up
pollution
and
depositing pollution
until it arrives back at its arrival
point.
This whole
procedure
for each of the 820
points
is
repeated
at
six-hourly
intervals 365
days
of the
year.
The model also
keeps
a record
of the
pollution produced by
each
country.
Not all the
deposition
can
be traced back to an identified
source,
as
meteorological
data is accurate
enough
to track back for
only
96 hours. Table 1
gives
a subset of the
basic data from this exercise for
1987,
and is to be read as follows.3
Looking along
the row
against
GB the table shows that Britain received
14,000
tonnes of
sulphur4
(i.e. about
27,000
tonnes
SO2)
from
France,
11,000
tonnes from West
Germany (DE)
and
571,000
tonnes from
domestic sources.
Looking
down the column headed GB the Table
shows that Britain emitted
1,271,000
tonnes of
sulphur
whose final
destination could be
established,
and of this
43,000
tonnes fell on
France.
45,000
tonnes on
Germany
and
437,000
tonnes on North
Africa within the
monitoring
area
(demonstrating just
how far the
plume
can
travel).
The
large
numbers on the
diagonal
of Table 1 shows how
important
domestic sources of
pollution
are. The
large off-diagonal
numbers
indicate where the
major impacts
of one
country
on another
occur,
and
it is
striking
that
they primarily
occur in East
Europe, confirming
the
view that central
planning
has been an environmental disaster for its
participants.
Table 2
presents
the information from the same
programme
in a
different
way.
The first two columns
give
total emissions
(not just
those
whose final destination can be
identified)
for the base
year
1980 and
the most recent
year available,
in order of
magnitude.
This allows an
estimate of the extent to which countries have succeeded in
moving
towards the
target
30% reduction now
widely accepted.
The next two
columns
give depositions
within the
country,
and an estimate of the
I 1
3
The fuller table is
given
in the
Appendix
and is the source of calculations
reported
below.
Using
the abbreviated table leads to considerable biases in
estimating
total
damage,
and the
complete
table should be used for all calculations.
4
To convert
sulphur
to
sulphur dioxide, multiply by 1.9,
or
roughly
double the numbers.
302 David
Newbery
Table 1.
Origins
of
sulphur deposition
in
Europe (thousand
tonnes a
year)
Emitters
CS FR DD DE BL HU IT PL ES SC SU GB OE UI Sum
~'
Czechoslovakia CS
France FR
GDR DD
West
Germany
DE
Benelux BL
Hungary
HU
Italy
IT
Poland PL
Spain
ES
Scandinavia SC
USSR* SU
Britain GB
Other
Europe
OE
N. Africa NA
Sum
Error
385 11 128 28 5 45 10 95 1
19 332 41 40 28 5 21 15 65
84 14 725 61 11 2 2 32 1
47 69 163 330 44 3 13 23 6
4 32 15 51 102 0 0 4 2
31 3 16 6 1 190 12 25 0
13 21 15 8 2 11 353 14 10
145 15 310 47 10 40 10 790 1
2 11 5 3 1 2 2 3 523
17 5 48 18 6 4 2 44 0
107 10 167 36 8 84 13 337 1
5 14 15 11 8 0 1 3 2
95 40 97 49 8 141 136 101 29
105 136 253 131 71 64 182 194 210
1,064
721
2,005
823 322 594 759
1,685
856
5 8 7 4 17 3 2 5 5
0 2 7 13 28 765
0 0 43 8 140 760
0 1 15 3 24 979
0 1 45 6 64 821
0 0 31 0 19 267
0 1 1 28 18 337
0 1 4 19 86 562
1 18 15 21 64
1,492
0 0 6 16 98 674
59 33 32 30 194 501
8
2,204
16 97 491
3,584
0 1 571 6 60 702
3 95 70 825 435
2,163
28 196 437 254 821
3,087
107
2,558 1,271 1,377 2,553 16,695
8 5 -22 51 11 1
Source: Acid
Magazine, Sept. 1989,
from EMEP data.
Notes:
Sulphur
dioxide
figures
will be about twice as
large.
*
European part
of USSR within EMEP area of calculation.
UI
= unattributable to
any country, plus
a small amount from N. Africa.
o0
coo
Receivers
Table 2.
Sulphur
emissions
(thousand
tonnes or
%)
Own contrib. Exports
Emissions
Change Depositions
1980 1987 %
1980 1987 total depositions Imports
USSR* 6,400 5,100
-20 5,101 3,584
0.61 2.1
GDR 2,500 2,500
0 963 979 0.74 7.0
Poland 2,050 2,270
+10
1,443 1,492
0.53 2.1
UK
2,335 1,840
-21 803 702 0.81 9.7
Spain
1,625 1,581
-3 670 674 0.74 7.0
Czechoslovakia
1,550 1,450
-6 818 765 0.50 2.8
Italy
1,900 1,252
-34 916 562 0.63 4.3
German Fed
Rep
1,600 1,022
-36
1,083
821 0.40 1.4
France 1,779
923 -48
1,160
760 0.44 1.4
Hungary
817 710 -13 416 337 0.56 3.5
Yugoslavia
588 588 0 662 497 0.39 5.0
Bulgaria
517 570 +9 293 235 0.65 5.0
Belgium
400 244 -39 162 121 0.41 2.7
Greece 200 180 -10 150 119 0.38 1.8
Turkey
138 177 +22 209 210 0.29 2.7
Finland 292 162 -44 273 210 0.23 0.7
Denmark 219 155 -13 110 83 0.37 2.4
Netherlands 244 141 -42 175 139 0.23 1.0
Portugal
133 116 -13 83 83 0.42 1.7
Sweden 232 116 -50 333 307 0.12 0.3
Romania 100 100 0 405 330 0.10 0.2
Ireland 110 84 -24 66 68 0.31 1.3
Austria 177 75 -58 282 207 0.09 0.3
Norway
70 50 -29 199 194 0.07 0.2
Switzerland 63 31 -51 121 70 0.11 0.4
Total
26,078 21,471
-18
20,484 16,695
0.52
Source: Acid
Magazine, Sept.
1989,
from EMEP data.
Notes:
Sulphur
dioxide
figures
will be twice as
large.
Countries ordered
by
1987 emissions.
*
European part
of USSR within EMEP area of calculation.
Figures
for emissions in 1987 based on
interpolation except
for
USSR, UK, Czechoslovakia,
Finland and the Netherlands.
fraction of
depositions
which can be traced back to domestic sources
(using
Table 1
data).
This confirms the
importance
of the
diagonal
element in Table
1,
and the
importance
of domestic sources of
pollution.
The final column
gives
the ratio of
exports
of
SO2
to
imports.
As total
depositions
in 1987 are
only
78% of total
emissions,
this ratio can be
expected
to be
significantly
above
unity
on
average.
The smaller it
is,
the more the
country
is sinned
against,
than sinner. Britain stands out
as the
greatest
sinner on this
criterion,
and the Scandinavian countries
as those most sinned
against.
Only part
of total
SO2
emissions come from man-made
sources;
other
important
sources include
volcanoes,
biological decay
and forest fires.
These natural sources
might
account for 80-290 mn. tonnes
per
annum
worldwide,
compared
to total man-made emissions of 75-100 mn.
tonnes. The levels and mechanisms
responsible
for natural emissions
are
imperfectly understood,
but
they may play
an
important
role in the
European
acid rain
problem.
The information
generated by
EMEP is
remarkably useful,
not
only
in
quantifying
the level of
pollution,
but also in
identifying
efficient
and feasible abatement
policies.
The information on
deposition
can be
used to draw
maps showing
the
average acidity
of
precipitation
over
Europe using
contour lines of
increasing
levels of
acidity.
Such
maps
show that in 1987 most of Yorkshire and the East Midlands had
precipi-
tation of
average acidity
below
pH 4.3, (i.e.
five times as acid as 'normal'
rain with a
pH
of
5.0)
whereas
Wales,
South-west
England
and the west
coast of Scotland was above
4.6,
and so less acid. Substantial areas north
of a line
joining
the Wash and
Liverpool
received more than 0.05
gm
H+
per square
metre
by
wet
deposition (i.e.
0.5
kg/hectare
or 50
kg/km2),
with
Wessex,
East
Wales,
Northern Ireland and North-east Scotland
receiving
less than 0.02
gm.
Most of the
sulphur deposition
in the UK
occurs
through dry deposition
as
SO2,
particularly
in the south and
east of the
country.
In the north and west where rainfall is more
frequent
and
intense,
wet
deposition
of SO2 and
sulphate particles
becomes more
significant.
In a
European context,
the lines of
equal
rainfall
acidity
show the
highest
concentrations in
Germany
and
Poland,
with
pH
below
4.1,
with most of Southern
France,
almost all
Italy, Spain, Portugal,
West
Yugoslavia
and West Greece
having
a
pH greater
than 4.9
(i.e.
less acid
than
any part
of the UK and
Ireland).
The EMEP tables can also be used to throw
light
on the
political
economy
of
pollution
control. Consider first the column in Table 2
which
gives
the fraction of total
deposition
which can be attributed to
domestic sources. The
unweighted average
of these
figures
is
41%
(with
a standard deviation of
21%).
The
weighted average (heavily
influenced
Acid rain
305
by
the
larger countries, especially
the
USSR, which, by
their
size,
have
higher
domestic
absorption)
is
52%.
The
weighted average
domestic
absorption
as a fraction of total
production
is
33%,
and the
unweighted
average
is
29%
(with
a standard deviation of
only 5%).
What this means
is that the
average
unilateral cost of
reducing
a tonne of domestic
deposition
is
equal
to the cost of
reducing
domestic emissions
by
about
3 tonnes. The ratio of total
European depositions
to total
European
emissions is
66%,
so that if all
European
countries acted in
concert,
the
cost of
reducing depositions by
1 tonne would be
only
half as
great.
Put another
way, many
countries could reduce
depositions
within their
borders
by
about
50%,
but at twice the cost
per
tonne reduced as if all
countries acted
together.
There are thus considerable benefits to coor-
dinated
action,
but these should not be dramatized -
SO2
pollution
is
far from a
pure public good
at the
country level,
and self interest
ought
to
go
a considerable
way
towards
alleviating
the
problem.
The next
question
one can ask of the EMEP data is whether there
are
significant opportunities
for bilateral
bargaining
between
pairs
of
countries over
pollution
levels. One
way
to
identify
such
opportunities
is to look for instances where the volume of bilateral
pollution exchange
is
large
relative to total
depositions,
and where trade is bilateral rather
than unilateral. The volume can be measured
by
one-half
exports plus
imports,
and bilateralism can be measured
by
the difference between
exports
and
imports.
Table Al of the
Appendix gives
the net ex-
ports
of each
country
and can be used to
identify
the extent of bilater-
alism. The
following country pairs
have a difference between these
two measures of 5% of
depositions
or less for the smaller of the
part-
ners:
Czechoslovakia-GDR;
Czechoslovakia-Hungary;
Czechoslovakia-
Poland; GDR-Poland;
Poland-Hungary;
USSR-Czechoslovakia;
USSR-
GDR;
USSR-Hungary;
USSR-Poland. It is notable that
significant
balanced
exchange
of
pollution
is confined to Eastern
Europe,
and
does not affect
any
of the other countries identified in Table 2.
(It
may
be substantial for smaller countries as a
proportion
of their
deposition.)
Another
possible question
to ask is which
pairs
of countries have
large
net trade balances in
pollution
which
might
lead to financial
negotiations
over
pollution
levels. The
following
countries have net
imports
from
another
country
which are
greater
than
5% of total
depositions:
Poland
from GDR
(19%);
Denmark from GDR
(12%);
Scandinavia from GDR
(10%);
Scandinavia from Poland
(9%);
USSR from Poland
(9%);
Czechoslovakia from GDR
(6%);
Scandinavia from USSR
(5%).
Scan-
dinavia thus contains the
only
West
European
countries which receive
large
net
imports
from
single
countries
-
otherwise it is the Eastern bloc
countries that stand out as
large
net
importers
from each other.
306 David
Newbery
Table 3. Emissions
per
head
(kilograms per
head
per year)
SO2 NO,
1980 1985 1980 1985
Austria 47 18 29 28
Czechoslovakia 202 203 78 73
France 65 31 34 29
Germany
Fed
Rep
52 43 49 49
Greece 83 73 13 15
Ireland 63 39 21 19
Netherlands 28 25 35 34
Poland 115 116 5 18
Spain
87 75 21 24
Sweden 64 33 38 37
Switzerland 19 15 31 33
UK 85 65 35 33
Canada 193 150 72 72
US 102 90 90 83
Averages:
All 86 70 39 39
Europe
76 61 32 32
West
Europe
59 42 31 30
Source: UNECE
(1987)
abatement.
National
strategies
and
policies for
air
pollution
Note:
Averages
are
unweighted.
Nitrogen
oxide emissions.
Nitrogen
oxide,
or
NOx,
emissions are measured
in terms of tonnes
nitrogen
dioxide
equivalent,
NO2.
Table 3
gives per
capita
emission levels for both
SO2
and
NOx
for the
major
member
countries of the UN Economic Commission for
Europe
Convention on
Long-range Transboundary
Air Pollution. It shows that UK levels are
not
high
in
comparison
with
Europe
and North America taken
together,
but are rather
higher
than the West
European
countries in the
sample.
The table shows that whereas
SO2
has decreased between 1980 and
1985, NOx
has if
anything
increased.
Table 4
gives
further information about
NOx
for 1985. The first
column shows that mobile sources contribute about one-half of all
NOx
emissions in Western
Europe (actually
OECD
Europe), though
the
range
is from
28%
in Eire to 84% in
Norway.
Much of the rest comes
from
large
combustion
plants-thus
in Britain 35% of the total came
from
power
stations. The next column shows total emissions of
NOx
from mobile and
stationary
sources in relation to total
energy
use.
The coefficient of variation
(CV)
is
34%,
showing
that emissions correlate
quite closely
with
energy consumption,
but there are
important
vari-
ations in the
degree
to which
energy
use causes
NOx pollution.
Britain
does
poorly by
this
score,
almost as
badly
as
Portugal
and Greece.
307 Acid rain
Table 4.
NO,
emissions from mobile and other sources
Ratio of total
NOx
to
Ratio of mobile
NOx
to
Mobile
NOx energy
(%
of
total) consumption
GDP car use petrol road fuel
Country (1) (2) (3) (4) (5) (6)
Austria 68 10.1 1.27 3.76 61 37
Belgium
55 9.0 0.84 2.73 44 22
Denmark 34 15.9 1.04 2.48 61 35
Finland 58 14.8 1.80 7.62 86 47
France 66 13.1 1.29 2.81 62 38
Germany
59 14.6 1.54 3.73 70 46
Greece 63 18.6 3.35 4.36 75 44
Ireland 28 10.7 1.08 22 14
Italy
51 14.8 1.06 2.03 70 34
Luxembourg
64 6.9 2.32 5.61 45 28
Netherlands 60 10.5 1.53 3.27 95 56
Norway
84 12.3 2.27 9.65 113 81
Portugal
38 19.6 3.38 4.13 123 54
Spain
46 18.5 1.49 4.50 67 36
Sweden 68 9.2 1.48 3.26 52 37
Switzerland 74 11.1 0.92 3.04 49 41
UK 45 16.4 1.65 2.96 44 32
Europe
49 14.1 1.42 3.15 63 39
SD
unweighted
14 4.76 0.83 2.01 25 14
CV 0.29 0.34 0.58 0.63 0.40 0.37
Sources: UNECE
(1987)
and OECD
(1987).
Data for 1985-86.
Notes:
(2)
shows
kg
of
NOx per
tonne of oil
equivalent
of
energy consumption
and
(3)
per
unit of GDP.
(4)
shows
gm
of
NOx per
km driven
by
cars.
(5)
and
(6)
show
kg
of
NO, per kg
of
petrol
and
kg
of total road
transport
fuel. SD is standard deviation and
CV coefficient of variation.
(Figures
for
Luxembourg
seem rather low and
may
be
explained by
some
energy
sales, especially
of
transport
fuel, being
consumed
abroad.)
Column
(3)
shows mobile emissions of
NO, per
unit of GDP
(which
has
a lower CV than total emissions
per
unit of
GDP).
Column
(4) gives
mobile emissions in
gm NO2 per
km driven
by
cars. White
(1982,
Table
2)
shows that if there were no emissions
regulations,
then for the US
64% of total
NOx
emissions would come from
cars,
and the balance of
36% from trucks. Uncontrolled emissions are 5.44
gm/km
for
cars,
and
38.6gm/km
from
heavy
diesel trucks. The
expected
uncontrolled
emissions
per
km driven
by
cars alone
might
therefore be 8.5
gm/km
(i.e. 5.44/0.64)
if the
proportion
of car km in total vehicle km were the
same as the US. The
European average
is 3.15 or
only
37% of that
predicted
for uncontrolled emissions.
Perhaps
more
impressive,
if the
308 David
Newbery
total emissions from all mobile sources are attributed
entirely
to
cars,
then the
average
achieved is about as
good
as those achieved in the US
by
cars of later than 1976 model
year (White, 1982,
Table
10).
The last
two columns relate mobile emissions to two
transport
fuels. Column
(5)
gives
mobile emissions in
kg per
tonne of
gasoline.
The final column
gives
total mobile emissions divided
by
total fuel consumed in the
transport
sector,
and thus accounts for diesel
emissions,
which are
potentially quite
serious.
Nitrogen
oxides also come from natural as well as man-made sources.
Again
estimates are
very imprecise,
but natural sources
may
account
for 20-90 mn. tonnes
compared
to estimated total man-made emissions
of about 90 mn. tonnes. One
might
therefore
argue
that
perhaps only
half the total acid rain emissions are
man-made,
but whereas natural
emissions are
worldwide,
man-made sources are concentrated in the
northern
hemisphere,
and
specifically
in
Europe
and North America.
2.3.
Assessing
the
damage
caused
by
acid rain
Acid rain has
ecological consequences
in that it affects the
soil, vegeta-
tion,
especially
forests,
lakes
(and
hence
fish).
It causes economic
damage
to man-made structures
(buildings,
fabrics, metals),
and it can affect
human health. The
ecological consequences
are
complex
and still
subject
to scientific
uncertainty
and hence
dispute.
Soils
vary widely
in their
ability
to buffer
(i.e. neutralize)
acid
rain,
and natural
processes
add to
the man-made sources of acid rain. Recent work undertaken for the
UNECE Convention on
Long-range Transboundary
Air Pollution5
attempts
to establish critical loads for various kinds of
soils, which,
if
exceeded,
would mean that the soil could no
longer
neutralize additional
acid rain
depositions. Many
sensitive areas
especially
in Scandinavia
experience
both
high
rates of
deposition
and soils for which critical
loads are low.
The effects of acid runoff on lakes have been
intensively
studied in
Scandinavia and the UK
(and
doubtless
elsewhere).
One of the main
mechanisms
leading
to the decline in fish stocks is the release of
aluminium caused
by
acidification,
rather than the direct effects of acid
(see e.g.
Environmental
Resources, Limited, 1983). Palaeoecological
studies of core
samples
can trace back
acidity
levels into the distant
past
and show substantial falls in
pH
in
many
lakes after the industrial
revolution. Thus Battarbee et al.
(1988) analysed
lake acidification in
sensitive areas in the UK and found that before 1850 most lakes studied
I I
5
Reported
in Acid News, No 3, October 1988.
Acid rain 309
had
pH
levels of about
6.0,
but that since then
pH
values had declined
by
0.5-1.5, (i.e. acidity
had increased
by
between 3 and 30
times)
depending
on
deposition
rates and
buffering capacities.
Other studies
show similar
trends,
and also show that lake acidification
may
be
reversed
(even
if
temporarily) by
the addition of lime either to the lake
or the rivers in the catchment area. This is
expensive,
as about 5 g/m3
of bicarbonate
(usually
in the form of
limestone)
are
required
to raise
the
pH
from 4.5 to
6.5,
and it does not
by
itself restore the lake to its
original
condition -
restocking may
also be
required. (See e.g. Dudley
et
al., 1985; Britt, 1986.)
One
implication
which does not seem to have been
adequately empha-
sized is that acid rain is a stock
pollutant
as well as a flow
pollutant.
That
is, part
of the final
damage
caused will
depend
on stocks of acid
in the
environment,
not
just
the rate at which acid rain is
deposited.
Even if the environment is
capable
of
neutralizing
or
disposing
of some
of the acid each
year,
if inflows exceed this rate of
disposal,
then the
stock of acid will increase. In the UK it is believed that current levels
of soil acidification are a
legacy
of the Industrial
Revolution,
and that
water
quality
will not be restored until the soil recovers. This
recovery
is a slow
process
and
relatively
insensitive to near-term rates of emission
reduction,
requiring liming
for
rapid recovery.6
This
may go
some
way
to
explaining
the
paradoxical relationship
between
decreasing
levels of
SO2
emissions on the one
hand,
and
apparently deteriorating ecological
conditions on the other. On the other
hand,
Battarbee et al.
(1988)
note
that acid
deposition
has been
declining
in Scotland over the
past
15
years,
and that the
uppermost
sediments are
already recording
an
improvement,
which
suggests
a
possibly
swift
improvement
if
deposition
levels could be further reduced. The
speed
of
response
will
presumably
depend
on the
ecological circumstances,
but
may imply high
rates of
'depreciation'
of acid stock levels.
Lake and fish
damage appears quite
well understood
compared
to
the
damage
suffered
by
trees. The
problem
was
highlighted
in
Germany
in the
early
1980s,
and shown to occur elsewhere. Forest
damage
has
been attributed to acid
rain,
weather
changes
and
droughts,
the
age
of
trees,
fragility
of soils at
high
altitudes,
and
inappropriate
forest
management.
Ozone attack
appears
to be
important,
and
may
have
synergistic
interactions with acid rain
(Environmental
Resources
Limited, 1983).
Even if the exact
proportion
of
damage
attributable to
acid rain is not
known,
there seems
widespread agreement
that reduc-
tions in acid rain would be beneficial to forests. Similar
uncertainty
pervades
the
study
of
crop damage, though again
ozone
appears
to be
I 1
6
Personal communication from Professor David Pearce.
310 David
Newbery
more
directly
harmful than acid rain. To the extent that ozone
plays
a
major
role in
crop
and forest.
damage, NOx,
which is a
major
contributor
to ozone
production,
is more
damaging
than
SO2.
Damage
to
buildings
and materials occurs
primarily
in urban areas
as a
consequence
of
relatively high
concentrations of
SO2,
with little
effect detected from
exposure
to
NOx.
The effects have been observed
and
correctly
attributed for
centuries,
and the estimated
damage
costs
are
thought
to be
high.
Health effects of intense
pollution
can be dramatic- as noted
above,
it is estimated that
4,000
people
died in the
great
London
smog
of
December 1952. Similar levels of
SO2
concentration were attained in a
subsequent episode
in London from 3-7 December
1962,
after the
Clean Air Act of 1956 had lead to a dramatic fall in smoke concentra-
tions. This time an estimated 340
died,
suggesting
that the earlier
episode
was so
deadly
because of
synergistic
interactions between smoke
particles
and
SO2 (Park, 1987, p. 127).
It
appears
that it is the
gas
SO2
that is
harmful,
rather than the wet form of acid rain. Acid rain in its
wet form can have indirect effects
by releasing
toxic
heavy
metals into
water
supplies.
Individuals
vary considerably
in their tolerance to these
gases,
but there is some evidence from
epidemiological
studies that
long-term exposure
at lower levels than these dramatic
episodes
can be
harmful to health
(Park, 1987, p.
127. See also Pearce and
Markandya,
1989,
for a
summary
of the extensive economics literature on the health
impacts
of
SO2.)
If
SO2
and
particulates
are
lethal,
the health case
against NOx
is at
best
unproven,
for it
appears
that
NOx
is much less active
biologically.
There is a certain
irony
in the fact that the
impetus
to
reducing
automobile emissions
initially
came from
California,
where it was
suspec-
ted,
and later established that
photochemical smog
was caused
by
vehicle
exhaust. The case mounted
by
the US Environmental Protection
Agency
for
reducing
emissions was based on the
supposed
adverse health effects
of
high
concentrations of
ozone,
though subsequent
studies
(Lave, 1982;
White, 1981; 1982)
cast considerable doubt on the evidence. To
quote
White
(1981, p. 59-60):
'... the ozone-related health effects under
discussion were short term and reversible.... Thus
far,
ozone
exposure
has not been demonstrated to have
long-term debilitating consequences
in humans.... The contrast with other studies of
pollutants,
such as
particulates
and
sulphates,
was
striking....
Particulates and
sulphates
probably
killed;
ozone
appeared
to do little more than cause
coughing!'
2.4.
Measuring
the costs of acid rain
damage
Estimates for the costs of different
types
of
damage
are scattered in the
literature,
and
vary greatly
in their
reliability.
Pearce and
Markandya
311
Acid rain
(1989) provide
a useful
methodological
discussion of cost-benefit
analy-
sis
applied
to environmental
pollution,
summarize a
variety
of these
estimates,
and note the criticisms to which
they
are vulnerable. One
approach
familiar to economists is to ask what
people
would be
willing
to
pay
for
property
located in less rather than more
polluted air,
as
reflected in the
response
of
property
values to
pollution
levels. Most of
the estimates here come from the 1960s and
1970s,
and
suggest
that
for each 1% increase in
SO2
concentration,
property
values fall
by
between 0.06-0.15 of
1% of their value for a house of
average
value.
An alternative
approach
is to look at the direct economic costs caused
by
the acid
rain,
and this has been done
by
for the Netherlands for
1986. It was estimated that current costs were about
$53-175
mn.
per
annum,7
but if the costs of
dealing
with future
damage
were taken into
account
(loss
of timber
etc.)
this
might
rise to
$120-380
mn.
Looking
at the current
costs,
the
large proportion
of the total comes from
agricultural damage,
thus extra
liming
of the soil to counteract
acidification
might
cost
$18-60 mn.,
and falls in
crop yield might
be
$36-360
mn. One should of course be most
wary
about
estimating
the
value of lost
agricultural output given
the distortions of the CAP.
Indeed,
as an
aside,
agriculture
is
responsible
for considerable
ground
water
pollution (notably
nitrates,
and
possibly algal blooms).
Much of
this is in turn the
consequence
of intensive
agricultural practices
induced
by
the
high agricultural prices enjoyed
under the
CAP,
notably high
fertilizer levels. If for various reasons it is difficult to reform
agricultural
output price levels,
then there is a
strong
case for
raising agricultural
input price
levels to the same ratio to world
prices
as
output prices
enjoy.
This would
improve
the
efficiency
of resource use and reduce
the
deadweight
losses associated with the CAP. Thus if
output prices
are twice
import parity
levels,
then fertilizer
prices
should be taxed to
raise their
price
to twice world market levels. This would
go
some
way
to
reducing
another form of environmental
pollution. (See Newbery,
1990, for the details of the
arguments
on efficient
input taxation.)
Other
ecological damage
estimates are
given
in Environmental
Resources
Limited, (1983).
German forest
damage
was
put
at
$0.25
bn.
p.a.,
and
rough
estimates of
potential
EEC wide
damage
can be deduced
from the annual value of
spruce
and fir
forestry production
of
$6.6
bn.
p.a.
Thus if
20% of forests are
adversely
affected so that their
production
drops by 10%,
the loss is
$0.13
mn.
p.a.8
OECD
(1981)
estimated that
I I
7
Unless otherwise
stated,
all cost estimates are in mid-1989
purchasing power, updating
from
the US consumer
price
index.
8
But note that there is still considerable
disagreement
as to how much forest
damage
has occurred
and how much is caused
by
acid rain.
312 David
Newbery
the value of fish loss in Scandinavia was
$38
mn.
p.a.
and in Scotland
might
be
$0.7
mn.
p.a.
Acid rain causes material
damage,
whose
costs, excluding
the costs
of
restoring
historic
buildings,
have been estimated
by
UNECE
(1982)
at between
$4-17/head (i.e. $1.0-4.7
bn. for the 1983
European
Com-
munity
as a
whole).
The
figures
in the Netherlands are somewhat
higher
($10-19/head)
and in
Germany
are estimated at
$19/head.
Environ-
mental Resources
Limited, (1983) gives figures
from OECD
(1981)
for
the estimated total corrosion
damage
for 12 OECD
European
countries
to
galvanized
steel and its
paint coatings
in 1974. For the UK the
figures
were
$5.9
bn.
p.a.,
for West
Germany
$9.5
bn.
p.a.,
and for
Belgium,
Luxembourg,
Denmark,
France and the Netherlands
together
$3.3 bn.,
or in total
$18.7
bn. How much of this can be attributed to acid rain is
still under
study.
Reducing
car emissions would also reduce
photochemical smog
in
some areas-
particularly
those which
experience temperature
inver-
sions combined with
strong
sun. Los
Angeles
is the
leading example,
but
clearly
Athens suffers
similarly.
There is no doubt that those
living
in such areas would be
willing
to
pay
for reductions in
smog
levels,
and
Schechter et al.
(1989)
estimate that households in
metropolitan
Haifa,
Israel,
would be
willing
to
pay
?12
(?1987) per
household
per
annum
to reduce
pollution
levels
by
50%.
It is difficult to
imagine
that this
would amount to a
large
total sum for
Europe
as a whole
compared
with the other
damage
costs,
given
the relative
infrequency
of
photo-
chemical
smog
in more
Northerly
climes.
2.5. The costs of abatement
The
Department
of Environment estimated the costs to the UK of
retrofitting
6 GW of coal fired
plant
with Flue Gas
Desulphurization
(FGD)
and all 12
major
coal fired
power
stations
(23 GW)
with low
NOx
burners at over ?1 bn. It now seems doubtful that more than a small
part
of this
programme
will
go
ahead,
as the
liability
to install FGD
would make the
privatization
sale of the CEGB unattractive. Instead it
appears
that the successor
companies
to the
CEGB,
Powergen
and
National
Power,
will meet the emissions standards
by
a combination of
installing high efficiency gas
turbines and
importing
low
sulphur
coal.
The
impact
of this on British Coal will be
substantial,
and it is in
interesting example
of how the
(private) cost-minimizing
solution to
the emissions standards
may
differ from
centrally imposed
solutions.
The Government also intends to
apply
the EC
large
car emission
standards 'as soon as
practical, probably
in the
early 1990s,'
at an
estimated annual cost of ?550 mn. The second
stage, applying
to small
Acid rain 313
Table 5. Estimates of costs of
reducing
SO2
by
various means
Cost
($1989) per
tonne of
SO2
Source Action removed
A
moving
to low
sulphur gas
oil
2,560
A
moving
to low
sulphur
fuel oil 640
A Fluidized bed combustors
(FBC):
new boilers 96
existing 2,240
A Flue
gas desulphurization (FGD):
new
plant
256
existing plant
640
B Drax FGD new
plant
350
C FGD retrofit
2,000
MW
plant
400-750
D 60% reduction from 40 GW CEGB coal
capacity
600
E FGD 90% removal 600
E FGD
marginal
cost of next 5% removal
1,600
F US coal
generators,
coal
switching,
Av cost 400
F US coal
generators,
no coal
switching,
Av cost 460
G East
Germany,
Wellman-Lord
FGD,
net of S sales 300
H move from 2.15% to 1%
sulphur heavy
fuel oil 380
H move from 1% to 0.7%
sulphur heavy
fuel oil 825
Sources:
A,
Environmental Resources Limited
(1983, p. 137, uprated by
1.28 to
$1989).
B,
Based on
Layfield (1987)
and
Jeffrey (1988). C, Longhurst
et al.
(1987). D, Dudley
et al.
(1985, p. 121). E, Brackley (1987). F,
Congressional Budget
Office
(1986)
in
Dowlatabadi and
Harrington (1989).
These are
average
costs.
Marginal
cost
might
be
twice
average
cost.
G,
Acid
News,
No.
3, July 1989, p.
9.
H,
Alfsen et al.
(1986).
cars,
was estimated to add an additional ?250 mn.
per annum,
a total of
about 4%
of UK
motoring
costs.
(Department
of
Environment, 1988,
7.14-15.)
This section examines various estimates of the costs of abate-
ment in somewhat more detail to see if
mandating particular
solutions
is
likely
to be cost
effective,
and to check on the
consistency
and
plausibility
of various estimates. The results are summarized in Table
5 and then
briefly explained.
The sources of these estimates are as follows. Environmental Resour-
ces Limited
(1983) gives
estimates of the
capital
cost of FGD at
$175-
200/kW
or about
15-20%
of the
capital
cost of the
plant. Retrofitting,
where
practical, may
increase this cost
by
a further
30-50%.
FGD
reduces thermal
efficiency by
about 2%
(e.g.
from
36%
to
34.1%)
and
so can increase the
operating
costs
by
10-20%. the first station which
the CEGB
plans
to retrofit is Drax A+
B,
which has a total
capacity
of
4,000 MW,
and which burns 11 mn. tonnes of coal
per
annum with a
sulphur
content of
1.7%. If
95% of this were
previously
released as
SO2,
the annual emissions would be
178,000
tonnes
S,
or
338,000
tonnes
SO2.
After
fitting FGD, 90% of the
SO2
will be
removed,
and the
314 David
Newbery
reduction in
SO2
would be
287,000
tonnes. The
Layfield report gives
the costs of FGD as
?17/kW/year,
broken down into ?5
capital,
?2.5
operating
and ?9.5 for loss of thermal
efficiency.
This last
figure
depends
on the cost of coal which has since fallen and
Jeffrey (1988)
estimates the
efficiency
cost as ?6.5.
Using
these
figures
the annualized
cost of the Drax
programme
would be ?56
mn.,
or about
?200/tonne
SO2
reduction.
Longhurst
et al.
(1987) gives
the CEGB's estimates of
the cost of
retrofitting
FGD for a
2,000
MW
plant
as ?160 mn.
plus
?35 mn. in lost
output,
or
?440-740/tonne
of
sulphur
removed from
the
gas
stream
(i.e. ?230-390/tonne SO2 removed). Dudley
et al.
(1985,
p. 121) try
to cost a
programme
to achieve a
60%
reduction in
SO2
in
UK emissions from its 40 GW coal-fired
capacity using
CEGB data and
data
presented
at the
Layfield enquiry.
The levelized lifetime cost of
sulphur
abated
ranges
from
?442/tonne
to ?738 with an
average
value
of
?550,
all in ?1983.
(The average figure
is thus ?745 in
?1989,
or
$1,200/tonne
S. The costs
per
tonne SO2 removed would be about half
this.) Brackley (1987)
estimates the cost of
SO2
removal
using
FGD as
$1,150/tonne
S removed with
90% removal,
rising
to
$3,000/tonne
S
for the next
5% removed in
going
from 90-95%
removal,
in both cases
using
coal with 1 %
sulphur
content. This
figure
is
very
close to
Dudley's
estimate.
The effect on the cost of
electricity generation
would be about
10-15%
of the cost of
generation
from coal-fired
power stations,
or
possibly
6%
of the
price paid by
customers.9 This can
usefully
be
compared
with
the
predicted
size of the 'nuclear
levy'
of
11%
of the sales
price,
which
will be
paid by
consumers of fossil-fuel
generated electricity
after the
privatization
of the CEGB to cover the cost of
supplying
20%
of total
electricity by
non-conventional
(mainly nuclear)
means.'0
Dowlatabadi and
Harrington (1989),
in a rather critical account of
US estimates of the costs of
large programmes
to reduce total emissions
by
8 mn. tonnes
per
annum from the current levels of 25 mn.
tonnes,
cites various estimates of the
average
cost
per
tonne
SO2
reduction.
Thus the
Office
(1986) gives
the least cost method
of
making
this reduction
(by allowing
the utilities to choose how best to
meet the
standards)
as
$360/tonne,
and
$400/tonne
if
they
must con-
tinue to use the same coal as
originally
instead of
substituting
to lower sul-
phur
content coal. These are
average
costs,
and there is considerable
I I
9
Based on estimated
capital
costs of 100-200/kWe
capacity,
5-7% discount
rate,
30-40
year
lifetime, 70% load
factor,
and 1.7%
sulphur
coal content. The calculated
figure
is
reassuringly
close to that
given
in 1984
by
the CEGB in HMSO
(1984, para 5.93).
10
This
suggests
the cost of non-conventional alternative is 55% of the final
price (11%
borne
by
80% of the total allocated to the 20%
non-conventional)
which seems
unreasonably high.
But
detailed estimates of the cost of nuclear
power
are not
yet
available.
Acid rain 315
agreement
that the
marginal
costs of
high
levels of
sulphur
removal
are
substantially higher
than small reductions. If the
marginal
cost were
twice the
average
cost at this
programme
level then the US
figures
would be close to the CEGB
figures.
Estimates based on German data
applied
to East
Germany suggest
that the
capital
cost
per
kW
capacity
is DM 600
using
the Wellman-Lord method." At
present
the
capacity
of the 15
larger plants
is 13.3 GW which is
responsible
for 1.98 mn.
tonnes of
SO2
p.a.
The
capital
cost would be DM 8
bn.,
and the estimated
annual cost of
operating
less the value of the
sulphur
sold on world
markets would be DM
1.6bn.,
or DM
600/tonne SO2 removed,
or
$300/tonne SO2,
which
appears
rather low.
An alternative
option
for
reducing
SO2
emissions is to switch to lower
sulphur
content fuels. Thus Alfsen et al.
(1986)
calculate the cost of
switching
from
high
to low
sulphur heavy
fuel oil as
2,300 NOK/tonne
SO2
removed
(i.e. $380)
when
moving
from 2.15% to 1% HFO,
and
5,000 NOK/tonne
removed
(=$823)
when
moving
from
1%
to 0.7%
HFO. Table 5 shows similar calculations of
switching
from
high
to low
sulphur
fuels for the EEC
given
in Environmental Resources Limited
(1983, p. 137).
What stands out from Table 5 is the wide variation in the costs of
the most common
proposed
method of
dealing
with
large power
stations- flue
gas desulphurization,
or FGD. In
part
this variation
may
be
explained by differing degrees
of
sulphur
removal - estimate E shows
that the
marginal
cost is
sharply increasing.
The East German estimates
may
be based on lower construction costs or a more
optimistic
view of
the value of recovered
sulphur.
The
figures
for FGD from source A
seem rather low when
compared
to other estimates.
Unfortunately
there
are no
European
estimates for the
important option
of
shifting
to low
sulphur
coal. In
part
this is
because,
unlike
oil,
there is no
clearly
defined
world market
price
for the two
grades
of coal that would allow a robust
estimate to be made of the differential cost of
shifting
from
high
to low
sulphur
coal.
The
study by
(1983)
concludes
that the estimated total
damage
caused
by
acid rain
might
be in the
range
$0.6-4.5
bn.
per
annum,
of which the
larger part
is
damage
to
buildings,
then to
forests,
then to
crops,
with fisheries
negligible.
The
costs of
reducing SO2
emissions in the
year
2000
by
10-27%
by adopting
FGD and FBC in new boilers
might
be
$1.9-6.5
bn.
per annum, depend-
ing
on the amount of new
capacity
installed. The cost of
retrofitting
might
be
$9
bn. to achieve a further 38-42% reduction. The cost of
d N , N. 3, Jy 1I
"l
Acid
News,
No.
3, July
1989.
316 David
Newbery
new FGD
might
be to raise the cost of
generating electricity
from coal
by
10-15%,
and the effect of
retrofitting existing
stations
might
be to
raise
generating
costs
by nearly
twice that. Since the
price
of
electricity
would be determined
by
the
marginal
cost of new
plant,
the effect of
forcing
older
plant
to
adopt
FGD would be to decrease their
profitability
and
equivalently,
to write down their
capital
value
sharply.
Once in-
stalled,
though,
their
operational
life should be the same as new
plant.
New
gas-fired generating capacity
will become
relatively
more attractive
than coal-fired
capacity,
as low
sulphur gas
is more
readily
available.
Recent
developments
which have
greatly
raised the thermal
efficiency
of
gas
turbines makes these cost-effective for base load at current
European gas
and coal
prices,
and
may
offer a lower cost alternative
way
of
reducing
acid rain emissions. The combined
impact
of more
stringent
emissions
standards,
movements towards a
single
market in
energy
with the
consequential
reductions in Government subsidies and
protectionist policies,
and the new
gas
turbine
technology may
sound
the death knell for the
European
coal
industry.
2.5.1. Costs of
reducing NO,
emissions. Estimates of the cost of
reducing
NOx
emissions are somewhat harder to
find, partly
because most studies
have concentrated on emissions from
cars,
where other
pollutants
were
also
being
reduced,
and
partly
because the
technology
for
removing
NOx
from
stationary
sources is not
yet commercially proven.
There are
extensive studies of the cost of
reducing NOx
emissions from vehicles
by
the use of
catalytic
converters and
engine
modifications, mostly
for
the US. It is
possible
to use the most recent to calculate costs
per
tonne
NOx
removed.
Using
the information
supplied by
Crandall et al.
(1986)
it
appears
that the costs of
introducing
the 1975 standards
compared
to no
regulation might
be
$5,500/tonne removed,
and the additional
costs of
introducing
the 1981 standards
starting
from the 1975 standard
would be
$9,000/tonne
removed
-
illustrating
the
rising marginal
cost
of abatement as emissions standards are
tightened.
Crandall et al.
(1986,
p. 114-15)
estimate that the
programme
costs for the US of the more
stringent
1984 standards
might
be about
$20
bn.
per
annum with a
replacement
rate of 10.5 mn.
cars,
which is several times rather
optimis-
tic estimates of the
potential
benefits of
reducing pollution. (Safety
regulations
in
contrast, though expensive,
seem to have been
justified
on cost-benefit
criteria.)
Another
way
to
put
the costs of emissions into
perspective
is to note that the
capital
cost of emissions
equipment
in
1981 was 6.7%
of the new car
price,
and the reduced fuel
efficiency
increases fuel
consumption by
5-15%.
The British estimate that
emissions controls
might
raise
driving
costs
by
about
4%
is consistent
with these
figures.
317
Acid rain
Recent
figures
from
Europe given
in IME
(1987) suggest
that
the cost of
meeting proposed
emissions standards are
$5,800-
10,600/tonne NO2 removed,
with lean-burn
engines
somewhat
cheaper
than
three-way catalytic
converters. This
compares
with costs of be-
tween
$200-4,000/tonne
removed from
stationary
sources
(using
modifications to burner
design)
and between
$4,000-10,000/tonne
removed
using (relatively untried) catalytic
converters. The best method
of
dealing
with emissions from
large
combustion
plants
is
disputed.
Selective
catalytic
reduction is
costly (perhaps
$150-200 per
kW
capac-
ity)
and
possibly
unreliable. Low-nox burners are
currently
under
development
and are
increasingly adopted
in new
plants. According
to
The Economist
(Dec 23, 1989)
there is a new
technology,
tested on a
150 MW
power plant
in West
Germany,
and in about 20 industrial
applications,
called selective
non-catalytic
reduction. The
technique
involves
injecting
a urea-based
liquid ('Noxout')
into the furnace
to reduce the
NOx
to
nitrogen.
The
capital
cost of
retro-fitting
is
low
($15-20/kW capacity)
and the cost of
NO2
removal is about
$1,500/tonne,
or less than
one-quarter
that of selective
catalytic
conversion.
3. Acid rain and environmental
policy
Acid rain is no
respecter
of national boundaries and has thus raised
international concerns. Debates about international
rights
and
obliga-
tions
appear
to be based on
legalistic
rather than economic
principles
-
of
equity, uniformity
and the
appeal
to
simple principles embodying
such
notions,
rather than to the more
finely adjusted
and
individually
variable notions of costs and benefits. We shall see first the outcome of
this international
debate,
and then contrast its
proposed
solution with
an
economically
rational solution.
3.1.
Policy
to date
The debate on acid rain and on
appropriate responses
has been conduc-
ted in two different forums. The initial
pressures
came from the UNECE
Convention on
Long-Range Transboundary
Air Pollution
(LRTAP)
which has 34 members from
Europe
and North America. Much of the
pressure
here was exerted
by
the Scandinavian countries and
Canada,
who are both
large
net
importers
of acid rain because of their unfortu-
nate downwind location relative to their
polluting neighbours
in
Europe
and the US. In
1982, Norway
and Sweden
pressed
for the
signatories
to reduce
SO2
emissions to 30%
below 1980 levels
by
1993. This led to
an informal '30% Club' founded in Ottawa in March
1984, and,
in
July
318 David
Newbery
1985 21
countries,
but not
including
the US and the
UK,
signed
a
protocol
at the third
meeting
of the UNECE LRTAP Convention in
Helsinki.
Whereas the Scandinavians were
initially primarily
concerned with
the acidification of lakes and streams and
consequent
loss of
fish,
West
Germans were worried about the
impact
their own
industry
was
having
on the
environment,
concerns which were reflected in the
growing
political power
of Green
parties
in the
early
1980s. An emotive
campaign
in 1982 drew attention to the
problem
of Waldsterben or forest
death,
in which official estimates showed that over half the forest area had
suffered
damage,
attributed to acid rain. For a
variety
of
political
reasons
described in more detail in Berkhout et al.
(1989),
a
Large
Combustion
Plant Ordinance
(Grossfeuerungsanlagen-Verordnung
or
GFAVo)
was
enacted in
June 1983,
under which 37 GW of coal fired
power
stations
would be fitted with flue
gas desulphurization equipment
and 12 GW
subjected
to
early
closure. Not
surprisingly, industry protested
that the
costs of this
programme,
which were to be borne
by electricity
con-
sumers,
would harm West
Germany's competitive position
in inter-
national
markets,
and this led the
government
to
press
for similar
standards
being adopted
for the whole of
Europe.
The
European
Commision
proposed
a
Large
Combustion Plant
(LCP)
Directive based
on the GAFVo in December
1983,
calling
for a cut in
SO2
emissions
by
1995
by
60%
to
40%
of their 1980 level. The Directive also
provides
for
stringent
emissions standards for new
large
combustion
plants.
In
November 1988 most
leading
industrial countries
signed
a UN
protocol
in Sofia to freeze the level of
nitrogen
oxides at 1987 levels until 1994
and
by
1996 to
agree
to further reductions based on critical levels.
The
proposed agreements
can be described as
aiming
at uniform
reductions in emissions from each
country
relative to an
arbitrary
starting
date,
and without
regard
to the costs and benefits at the
country
level.
3.2. The economics of
designing
a
policy
for acid rain
Welfare economics has
developed
a
range
of theories to deal with
externalities and
public good (or bad) problems
such as acid
rain,
which
hardly
need
rehearsing
here. Acid rain differs from
'greenhouse gases'
like CFCs and carbon dioxide in that it is not a
pure public good (or
bad)
in the Samuelson sense. Acid rain causes
damage
where it is
deposited,
and there
appears
to be a
reasonably
linear
relationship
between emissions and
depositions,
as Table 2
suggests
for the
aggre-
gate. Moreover,
the EMEP is
predicated
on there
being
a
predictable
and stable
relationship
between the location of the source and the
Acid rain 319
deposition,
at least
averaged
over a
year.
Detailed
work, reported
in
Derwent
(1988, 1990), suggests
that both
SO2
and NOx
depositions
can
be
statistically predicted, given
the location of the
source,
and that
depositions
at
specified
sites can be traced back to their
originating
source. The main differences between the two
pollutants
is that a
higher
fraction of
SO2
is
deposited
within 100 km of the
source,
and that a
larger
fraction of
SO2 can be accounted for
by individually
identified
large stationary sources,
predominantly power
stations.
Nitrogen
oxides
are
produced
both
by large stationary
sources and
by
numerous small
mobile sources - road vehicles.
Technically,
acid rain
appears
to be a
depletable
or rival
good
in
consumption,
in that if 1 tonne of
sulphur
falls on a
given
local
area,
then that tonne cannot fall
elsewhere,
and reduces the amount which
will harm others
by
that amount. It
might appear
to be a
simple
bilateral
externality
of the kind considered
by
Coase
(1960).
But there are two
important
differences from the
simple
case of bilateral
externality
in
which a well-defined
polluter deposits pollution
on a
single
well-defined
recipient.
On a
given day,
the
plume
of
pollution
from a source can be
tracked,
and the amounts of acid rain
deposited along
its
flight path
estimated. The
plume may deposit pollutants
on a number of different
countries. Even over a short
period
of
time, then,
more than one
recipient
will be affected
by
the
single pollutor.
Over
longer
time
periods
the number of
recipients
will be
larger
as the wind direction and
strength
vary,
even if the
average deposit
received
by
a
given country
from a
given
source is well-defined over some time
period.
If the
recipients
are to
bargain
over reductions in acid
rain,
they
will have to
agree
among
themselves how to coordinate their
bargaining,
and how to share
any
costs involved.
The second
complicating
fact is that
reducing
emissions
requires
a
fixed durable investment in FGD
equipment,
as well as variable costs
(largely
additional
energy
costs to
operate
the
FGD).
Even if there were
only
one
recipient
at
any
date,
there will be a number of
recipients
over the life of the FGD
equipment,
all of whom benefit from its
installation.
Again,
there are
joint beneficiaries,
and the FGD
plant
itself is like a
public good,
available to all
potential recipients
once
installed.
Third,
there are several alternative
ways
of
reducing
the
damage
caused
by
the main source of
SO2, power
stations. One is to
reduce emissions
by
investment in
FGD,
another is to reduce emissions
by reducing
the amount of
electricity generated by
a
given
station. In
an
integrated electricity grid, power
stations are ranked
by
merit
order,
with the
cheapest
variable cost
plants given
the
highest rank,
and the
most
expensive
the lowest. The
despatcher
calls for
power
from the
highest
ranked stations and moves down the merit order until total
320 David
Newbery
demand is satisfied. On a
given day
the wind direction and
strength
may
cause the
plume
from one
power
station to
deposit
acid rain where
it does
high damage,
while that of another
power
station
may
fall
largely
on the sea or on areas where
damage
costs are low. If the
power
despatcher
includes in the costs of
generating electricity
the social cost
of the
damage
done
by
emissions,
and if these costs are varied with
changing
weather
patterns,
then the
position
of the
power
station in
the merit order will
change
with the
weather,
and the
damage
done
by
the
pollution
will be
reduced,
even if the total amount of
pollution
remains
roughly
the same.12 Much the same result can be obtained
by
holding
stocks of
low-sulphur coal,
which are burned when the wind
direction is adverse in
preference
to the use of normal
(higher sulphur)
coal.
Finally,
if the
public good
and coordination
problems facing
the
recipients
of
pollution
from a
given country
can be
overcome,
then it
should be
possible
for these
recipients
to
bargain
towards an efficient
level of
pollution,
as
argued by
Coase
(1960).
How this
might
occur is
discussed below.
A number of
very important
economic
consequences
flow from these
facts.
First,
the
damage
done
by
acid rain
depends critically
on where
the
deposition
occurs. If it occurs over the
sea,
it is
unlikely
to have
much harmful
effect,
as the sea is
naturally
alkaline. If it occurs over
sparsely populated
areas with class 5
(acid tolerant) soils,
then
again
the
damage may
be low. If it falls on rivers and
lakes,
it
may
be
very
damaging ecologically,
but
financially
the costs look trivial
compared
to
property damage.
If it falls on
densely populated areas,
the
property
damage
will be substantial. As a
consequence,
the benefits of
reducing
emissions will
vary significantly
from source to
source,
since different
sources will have
very
different
deposition patterns. Second,
the costs
of
reducing
SO2
emissions will
vary significantly
with the source. The
cost
per
tonne
SO2
reduction will
depend
on whether an
integrated
FGD
system
can be installed at the time of construction of a new
station,
or whether it must be retrofitted. If it is
retrofitted,
the cost will
depend
on the number of GW hours the
plant
will
produce
over the remainder
of its life. For a base load
plant
with a
long expected life,
the cost will
be
low,
but for a
plant
near the end of its
life,
or one which is
primarily
used for
peaking,
the cost will be
high.
If the
object
is to maximize the
Europe-wide
net benefit of
pollution
reduction,
then three
types
of
relationships
are needed. EMEP divides
12
Estimates
presented by
Sir Sam Edwards of the Cavendish
Laboratory
in
Cambridge suggest
this
may
be a
very
cost-effective
way
of
reducing pollution
costs for UK
generators.
Acid rain 321
Europe up
into 150 km
square
cells and measures
depositions
in each
cell. This is
probably
be the finest
degree
of
disaggregation practical,
at least in the medium run. For each of these cells one needs a
damage
function,
relating
total
damage
to
depositions.
Let
MDi
denote the
marginal damage
when one extra unit is
deposited
at location i.
Suppose
tij
denotes the fraction of
pollution produced
in location
j
which is
deposited
in location i.
Finally,
let
MCj
denote the
marginal
cost incur-
red in location
j
from
cutting
back or
abating pollution by
one unit in
location
j.
Then total social costs will be minimized when for all locations
i and
j
MCj
=
tiMDi
For each location
j,
the left-hand side is
simply
the
marginal
cost of
pollution
abatement,
and the
right-hand
side the
marginal
benefit of
pollution
abatement, namely
the avoidance of the
marginal damage
which another unit of
pollution
would
bring
about in all locations.
Detailed
figures
on the
damage
functions are not
readily
available,
but it is
interesting
to consider the
implications
of three different
measures of
damage.
First,
we need to measure the
intensity
of
pollution
produced by
a
given
source,
and the natural measure is
SO2 per square
mile of the
recipient country per
tonne emitted. The total
damage
done
will then
depend
on what is
damaged.
Given the dominance of
property
damage
one
possibility
is to make the
marginal damage proportional
to
capital
stock times
pollution intensity. Property
(and
possibly
other
relevant
components
of
damage)
are
likely
to be
proportional
to
GDP,
and so if we
multiply
tonnes
deposited per
tonne emitted
by
GDP
per
square
mile,
then we have a
rough
measure of
property damage.
Given
the
unreliability
of estimates of GDP from socialist
countries,
the
measure of GDP
finally adopted
are the
purchasing power parity
measures for 1985
given
in Summers and Heston
(1988).
Estimates of
the
geographical dispersion tij
can be derived from the fuller version
of Table
1,
given
in the
Appendix.
It is more useful to
compute
an
index of these costs and Table 6 below
gives
these
figures
for various
source
countries,
ranked
by
the
damage they export,
relative to Britain.
The index of
property damage
is the column headed
'prop'.
An alternative measure
might
be the number of
population
in the
country affected,
again multiplied by
the
pollution intensity.
This index
is
given
in the column headed
'Pop'. Finally,
as a measure of
willingness
to
pay
for reductions in
pollution,
the third measure
multiplies pollution
intensity by
GDP
per
head,
and is
given
in the column headed 'GDP'.
Four countries
appear
to be as or more
damaging
than the UK
by
the
property damage
and
by
the
population
criterion,
and eleven
by
the GDP
measure, though
one should remember that the estimated
322 David
Newbery
Table 6. Index of
damage per
tonne
SO2
emitted
Index of
damage
Reduction %
Source
prop
pop
GDP
prop
pop
GDP
Belgium
175 160 149 39 30 26
Netherlands 164 154 121 65* 40 18*
GDR 156 139 155 83 80 75
France 100 92 146 29 28 28
UK 100 100 100 60 60 40*
FRG 95 100 136 36 28 23*
Luxembourg
84 69 88 6 6 6
Switzerland 78 68 97 25 25 25
Czechoslovakia 68 80 116 55 54 52
Italy
65 76 82 39 29 22
Denmark 46 41 111 31 21 35
Hungary
46 69 104 60* 65 70
Austria 45 51 92 30 30 30
Poland 43 65 95 26 33 33
Yugoslavia
35 55 77 63 75 75
Spain
30 37 69 10 11 36*
Ireland 27 33 51 11* 11* 12*
Portugal
25 49 53 10* 40 30*
Bulgaria
18 33 50 17 26 27
Greece 13 27 38 10* 60* 60*
Sweden 10 9 109 0 0 3
Albania 9 26 17 0 10 0
Romania 8 20 45 0 7* 15*
Turkey
8 25 30 0 0 0
Norway
7 5 104 0 0 6
Finland 7 7 105 0 0 18
USSR 5 6 73 2 3 20
Average
54** 59** 89** 30*** 30*** 33***
Notes:
Damage
is
deposition per
tonne emitted times
damage measure,
which is
GDP/sq.
mile
(prop), population,
or
GDP/head, taking damage by
UK as 100. Countries are
ranked
by
total
damage using
the first measure of
damage.
Reductions are those
required
to achieve an overall reduction of 30% at minimum cost
according
to the
three measures of cost.
Damage
index for USSR are related to
population
densities in
the EMEP area alone. *
estimate unreliable as MC of abatement flat.
**
unweighted
average.
***
weighted average.
damage
has been
very crudely
estimated. As
SO2
does not travel
very
far,
and as the
damage
is
weighted by
the GDP
density,
countries like
Belgium
which are situated near
wealthy neighbours yield high damage
costs
per
tonne
SO2
released. At the other
extreme,
the
USSR,
which
produces
a vast amount of
SO2,
deposits
most of it within its own
borders,
and even here the
damage
cost is low because of the low
population
and GDP
density.
It is
interesting, largely
as an illustrative
exercise,
to see what
kinds of
SO2
reduction would be
justified given
the
marginal
costs of
Acid rain
323
324 David
Newbery
abatement estimated
by
Amann and Kornai
(1987).
That source
gives
plots
of total and
marginal
abatement costs
by tonnage
abated
by country
for the
year
2000
(i.e. allowing adequate
time to undertake
adjustments,
and
predicting
future levels of emissions without
abatement).
These
plots
have been used to determine
marginal
costs of abatement and the
levels of reduction which would be
justified-the
level at which the
marginal
cost of abatement is
equal
to the
marginal
damage
which
would be
appropriate
for a
30%
overall reduction in emissions. If we
had absolute measures of the
damage
then it would be
possible
to
determine the
appropriate
total and individual reduction which is
cost-justified,
but as we
only
have an index of relative
damage,
this
alternative finds the least-cost method of
making
the
30% reduction.
Table 6
gives
the
implied
reductions which minimize costs. In some
cases the estimate is
reasonably
robust,
assuming
the correctness of the
underlying
abatement cost
figures.
In other
cases,
indicated with a
single
star,
the
marginal
cost of abatement was flat at a level close to
the estimated
marginal damage,
and so the
appropriate
level of reduc-
tion is
poorly
identified.
Figure
1 illustrates the
problem
for the UK.
When the
damage
is measured
by
the third measure based on
ability
35 -
30 -
C
c
Marginal
cost of
0
25 -
abatement
0.
o 20 -
0
o
15
-
n
0 o
-10
-
Measures of
.C t-
marginal
benefit
5 -
j
/
Prop damage
.. .... . ....... .. .. ..
Pop
dam
age
?7..
"?Z
--?---GDP/hd
damage
0
I I I I l
0 1 2 3 4 5
Abatement (million
tonnes
S02)
Figure
1. Abatement costs for the UK
Source: Amann and Kornai
(1987).
to
pay (GDP/head)
the
marginal damage
and
marginal
costs almost
coincide over a
range,
whereas for the other cases the
marginal damage
intersects the MC schedule at a vertical
step.
The reductions
proposed vary significantly by country
and also
according
to the measure of
damage. They
are not
simply
related to
the
damage
index,
as abatement costs
vary considerably by country.
Thus
although Luxembourg
has a rather
high damage index,
the
marginal
abatement costs are so
high
that
very
low reductions are
justified.
In some cases the abatement costs of further reductions are
high
relative to the benefits of the reduction and for these countries
zero reduction is
proposed.
One obvious conclusion is that a uniform reduction in
pollution
from
an
arbitrary
date like 1980 is most
unlikely
to maximize social net
benefits.
Instead,
abatement should take
place
in an order determined
by
the size of the excess of social benefits over abatement
costs,
continu-
ing
until the net benefit falls to zero. In the illustrative case shown in
the
table,
Britain
appears
to warrant a more radical abatement
pro-
gramme
than most of her EC
partners, primarily
because abatement
costs are
quite
low and
damage
costs
appear
to be
high.
It should be
stressed that this conclusion is
only
as
good
as the data on which it rests
and the
methodology
for
estimating damage
costs.
The exercise
reported
in Table 6 makes
strong assumptions
about
costs and
benefits,
and it is therefore
important
to look at other similar
exercises based on an economic
approach
to costs and benefits. Maler
(1989)
has estimated cost and
damage
functions for the same set of
European
countries and used the 1984 EMEP transmission matrix to
compute
the efficient level of abatement. He assumed the costs of
abatement were
quadratic (i.e.
the
marginal
costs were
linear,
which is
a
poor assumption
for
many
of the
countries).
As in Table 6 the cost
schedules were calibrated to cost estimates made at IIASA for the
year
2000
using
the schedules in Amann and Kornai
(1987).
He also assumed
that
damage
costs were constant
per
tonne of
sulphur deposited
in each
country
(as
was done in Table
6),
but that the
damage 'represents
the
evaluation of the
damage
that
respective governments
make
today'.
These estimates are made as follows. If each
country
takes the level of
emissions of all other countries as
given (the non-cooperative solution),
and if each
country
then balances the
marginal
cost of domestic abate-
ment with the
marginal
benefits of reduced domestic
deposition,
then
it will solve the
equation
MCi
=
tii MDi
If
MCi
and
tii
are
known,
then
MDi
can be deduced. On the
strong
assumption
that the
governments
of
European
countries were
rational,
Acid rain 325
Table 7. Net Benefits from an efficient solution
% abatement
(various measures)
net benefit
prop pop
GDP/hd
Maler (DM
million)
GDR 83 80 75 80 11
Netherlands 65 40 18 62 565
Yugoslavia
63 75 75 79 344
Hungary
60 65 70 77 5
UK 60 60 40 81 -365
Czechoslovakia 55 54 52 75 152
Belgium
39 30 26 36 191
Italy
39 29 22 33 -81
FRG 36 28 23 86 328
Denmark 31 21 35 86 119
Austria 30 30 30 21 324
France 29 28 28 10 879
Poland 26 33 33 27 599
Switzerland 25 25 25 23 192
Bulgaria
17 26 27 43 -7
Ireland 11 11 12 38 71
Spain
10 11 36 14 -29
Portugal
10 40 30 19 10
Greece 10 60 60 86 53
Luxembourg
6 6 6
USSR 2 3 20 2
1,505
Albania 0 10 4 45 24
Romania 0 7 15 83 422
Finland 0 0 18 14 -2
Norway
0 0 6 6 272
Turkey
0 0 0 9 68
Sweden 0 0 3 4 606
Total 30 30 33 39
6,290
Sources: Maler
(1989)
and Table 6 above.
Note: Net benefits as estimated
by
Maler.
selfish and
non-cooperative,
the revealed
marginal willingness
to
pay
for abatement can thus be deduced.
(Maler, 1989; 1990,
p. 90.)
His estimates of the efficient level of abatement involves an
aggregate
reduction of 8.9 mn. tonnes
sulphur,
or a reduction of
39%,
yielding
a net benefit of DM6.3 bn. The main conclusion to be drawn from this
estimate is that the efficient solution
requires very
different reductions
in each
country,
and that a uniform
percentage
reduction would
yield
much lower social benefits. His basic table
gives
the
percentage
reduc-
tions and net benefits for selected countries shown in Table 7. For
comparison
the
percentage
reductions for the three different measures
of
damage
from Table 6 are
presented
next to Maler's estimates. The
relation between the different
proposed
reductions is also
graphed
in
the scatter
diagram
in
Figure
2. In
interpreting
the
figure
it should be
326 David
Newbery
Acid rain 327
100 -
x x x Population
x
80 - x
x
x
x
X + + GDP/head
+
.0-
x
x Maler
x
5 60 - +
L
+
E x
x
x
X 40
- * +
'
~0
+
x + x
20
X
+
x
x-
++
2
+
x
+
0
dI I I I I
0 20 40 60 80 100
Property damage
reduction %
Figure
2. The relation between
proposed
reductions
remembered that Maler's estimated overall level of reduction is
39%,
rather than the
30% used in the calculations of Table
6,
so that his
figures
will be on
average higher.
It is
interesting
(and
rather
surprising,
given
the
very
different
approach adopted)
that the order and the
degree
of
pollution
reduction in Maler's calculation is not so different
from those
appropriate
to some measures of Table 6. Table 8
gives
the
correlation matrix between the various
proposed
reductions. As
expec-
ted,
Maler's measure has a lower correlation with those
proposed
here
than each of the three measures have with each other.
Table 7
shows,
on Maler's
figures,
that not all countries are net
beneficiaries of the efficient
plan,
and the losers
might
not
agree
to
join
the
agreement.
The fact that most countries
gain
from a move to the
efficient
plan
is also rather
misleading,
for if countries have to
agree
to the
percentage
of the efficient level of
joint pollution
reduction to
undertake,
then
many
countries will lose from the final increments -
say,
in
going
from 90-95%
of the
plan.
Maler
argues
that in a
repeated
game
with
appropriate
side
payments, supporting
the efficient solution
could be made
individually
rational
-
deviating
countries would
trigger
David
Newbery
Table 8. Correlations between different
proposed
reductions
(R2 percentages)
Prop Prop
GDP/head
Pop
72
GDP/head 52 82
Maler 38 46 39
punishments
in the form of increased
depositions,
and would lose
any
side-payments.
3.2.1.
Bargaining
over abatement. Let us return to consider the
efficacy
of
bilateral or multilateral
bargaining
over
pollution abatement,
in the
spirit
of Coase. One can
imagine
two
possible
allocations of
property
rights.
The status
quo
is one in which each
country
is free to
pollute
its
neighbours.
The natural alternative is one in which each
country
agrees
to a certain annual level of emissions -
e.g.
70%
of the 1980
measured level. Emissions above that level are
acceptable only
with the
agreement
of the
recipient
countries. In the second
case,
the
polluter
would have to
pay
for increased
emissions,
and a natural offer would
be an amount between the
marginal
cost of abatement and the
marginal
damage
done to the
recipient.
This
principle
is termed the Polluter
Pays Principle
or PPP. The
important question
of how this should be
decentralized within the
polluting country
will be considered
below,
but in
principle any
new
pollution
source would have to
buy
the
right
to release a
given
number of tonnes of
pollution
from an
existing
source,
which would then reduce its emissions
by
that amount.
In the status
quo situation,
new
polluting
sources could set
up
and
pollute
at no
apparent
cost. The costs of
damage
would be borne
by
the
recipients (the
Victim
Pays Principle,
or
VPP).
Does this mean that
the status
quo
would favour
entry by
more
polluters? Ignore
for the
moment the
problem
of coordination
among
the
recipients,
and
suppose
that there is one
representative recipient
who is
willing
to
bargain
with
the
polluting country,
and one
polluter
within the
country,
such as the
CEGB in Britain.
Suppose
also that there are no market failures in
selling electricity,
of the kind considered in the next section. Then it
is a standard Coase result that the allocation of
property rights
does
not affect the
efficiency
of the final outcome. The first task would be
to
agree
a reference
path
of
pollution
in the absence of
any cleanup
-
presumably
based on a forecast of
energy
demand and emissions
per
GWhr. The
recipient representative
would then offer to
pay
the CEGB
for reductions below this level. In both cases the social cost of
producing
328
electricity
would be the same and would include the social cost of
releasing pollutants.
In the first case
(PPP)
the CEGB would
pay
the
recipient
for extra
emissions,
and in the second case
(VPP)
the CEGB's
total
profits
would decrease for each extra tonne released - the cost of
emissions would be the reduction in
profit.
There is a further
similarity
in the
operation
of the two forms of
property right,
in that
recipient
countries
may
find the benchmark
figure
of
70% of 1980 levels too
high,
in which case that becomes the
new status
quo point.
Economic
growth
which raises the wealth
density
is
likely
to make further absolute reductions
cost-effective,
especially
as technical
progress
should continue to lower abatement costs. The
recipient
would then wish to
pay
for reductions below the status
quo,
just
as before. The
only
difference between the two cases would be in
the size of the
lump-sum
transfer from
recipient
to
polluter.
In the
second case the
recipient
would
pay
the
polluting country
the
damage
cost of the difference between
70% of 1980 levels and the reference
emission
path.
3.2.2.
Complications
caused
by regulation
and
privatization. Suppose
the elec-
tricity supply industry (ESI)
is
privately
owned but
regulated,
as in the
US. If
regulation
is based on actual costs
(again
as in the
US),
with the
allowable
price
related to the
average
cost,
then the
average
cost of
electricity
is lower under VPP than under
PPP,
and hence
prices
will
be
lower,
demand
higher,
and the level of
pollution
will be
inefficiently
high.
The
appropriate
method of
price regulation
would be to deter-
mine benchmark costs based on a
pre-determined
level of emissions
control,
and allow the
generators
to
keep payments
for reductions and
force them to
pay
for emissions above the status
quo
level.
Naturally
this
particular problem
does not arise with
price-based regulation
of
the RPI-X
typified by
the
regulation
of British Gas or British Telecom.
The
present proposals
for
privatizing
the
electricity supply industry
in the UK involve
splitting
the CEGB into three
components.
All nuclear
power
stations will remain in
public ownership
(and
produce
no acid
rain
pollutants anyway).
About two-thirds of the
remaining capacity
will be allocated to National
Power,
the rest
going
to
Powergen.
The
grid
will be owned
by
the 12 distribution
companies (Discos).
New
entrants and
existing private power suppliers
will be free to sell to the
grid
or Discos. The
price
of
electricity
sold
by
the Discos will be
regulated,
but not that sold
by
the
generators,
who will be
subject
to
normal market
competition.
Suppose
that the British
government accepts
a
target
level of abate-
ment
(e.g.
70% of 1980
levels),
and must decide how to
implement
this
target.
The two standard instruments available are emissions taxes
Acid rain 329
(per
tonne of
sulphur
or NOx
emitted),
or standards
(e.g.
tonnes
SO2
emission
per GWhr).
Standards
applied uniformly
are
typically
inefficient,
and economists
usually argue
that their
inefficiency
can be
eliminated
by auctioning
a fixed number of licenses to
pollute
1 tonne
SO2
in a
given year, equal
to the total allowable emissions. If
polluting
firms are
competitive,
and the tax is
appropriately
set
(at
the auction
price)
then standard
arguments
show these two methods to be the same
under
certainty.13
But the British ESI will not be
competitive,
at least
in the near future until considerable
entry
takes
place,
and so this
argument
does not
apply.
In the
Appendix
a
simple
Cournot
duopoly
model is constructed to see how an auction market in licenses to
pollute
might
work.
The model shows that if firms
compete
in the
input
market
(for
licenses to
pollute)
as well as in the
output
market,
then it
may pay
them to raise the cost of
inputs
to each
other,
as a
way
of
increasing
the final
product price.
The idea that
raising
rival's costs
may
be a
preferable competitive strategy
to
predatory pricing
is not
new, though
the usual
assumption
is that firms are
differentially
affected
by input
costs
(see e.g. Salop
and
Scheffman, 1983).
The normal
argument
is
informal,
if
intuitive,
and
ignores
some of the subtleties in
modelling
competition
in both
input
and
output
markets. The model in the
Appendix
considers two identical firms which can reduce emissions at
a
cost,
and which are allocated or bid for licenses to
pollute.
If
they
behave as Cournot
duopolists
in the
output
market,
then their
joint
profits
will be
higher
under an
asymmetric
allocation of licenses than
with the efficient
symmetric
allocation. The reason is that an
asymmetric
allocation raises
marginal production
costs of the smaller firm and allows
the two firms to
compete
less
intensely,
to their
joint
benefit. What this
implies
is that there is a
mutually
beneficial transfer
price
for licenses
between the two firms which
compensates
the firm with the smaller
number of licenses more than its fall in
profits. Unfortunately
the
increase in
joint profits
is smaller than the increased
deadweight
loss
of the inefficient allocation of
licenses,
so net social
surplus
decreases.
Consequently,
taxes and
quotas
or auctioned licenses are not
equiv-
alent in the
presence
of
imperfect competition
- for taxes can in
prin-
ciple
mimic the
competitive
market
provided
the
government
knows
that abatement cost function and can set the tax level
appropriately.
There are additional
arguments differentiating
licenses and taxes.
If there is
uncertainty
about the location of the
damage
and abatement
cost
functions,
then it
may
be
preferable
(in
the sense of
lowering
the
I I
3
Just
as tariffs and
quotas
are
equivalent
under
certainty, though
not under
uncertainty (see
Newbery
and
Stiglitz, 1981).
330 David
Newbery
expected
social
cost)
to choose one instrument rather than another. In
Appendix
B taxes are shown to dominate
quotas
for a
competitive
industry
in which there is
uncertainty
about the
slope
and level of the
marginal
cost of abatement schedule and
similarly
about the
slope
and
level of the
marginal damage
cost
schedule, provided
the
expected
slope
of the
marginal
cost of abatement exceeds the
expected (absolute
value of
the) slope
of the
marginal damage
schedule. This seems
plaus-
ible,
for the
damage
schedule is
likely
to be rather flat if
damage
is
proportional
to
depositions,
while the
age
distribution of
large power
plants
will introduce an
upward slope
into the abatement schedule.
The
proportional
relative
advantage
of taxes over
quotas
will be
(b
-
d)2/(b+ d),
where b is the
slope
of the abatement
schedule,
-d is
the
slope
of the
damage
schedule,
and ao is the coefficient of variation
of the
slope
of the abatement
schedule,
a measure of the
uncertainty
about the
marginal
costs of abatement.
3.2.3.
Choosing policy
instruments. Licenses to
pollute
which are tradeable
and are auctioned off have several attractions. The first is that
they
are
the natural instrument to meet international
agreements
couched in
terms of total emission
levels,
typical
in recent
negotiations.
Second,
they
can overcome
organized
resistance from the
industry
affected,
since firms can be allocated licenses
proportional
to current emission
levels. The costs of abatement fall on consumers
(who forgo
the revenue
from
auctioning
off the
licenses),
and on new
entrants,
who have to
buy
licenses from incumbents. This
anti-competitive
feature has
already
been
discussed,
and is a reason for industries
favouring stringent
stan-
dards with
appropriate grandfather
clauses.
Third,
other countries can
negotiate
further reductions
by buying up
licenses,
providing
the
issuing
authority recognizes
their
property rights
in such licenses.
The limitations of licenses are that
they
allow
anti-competitive
behaviour,
they
are inferior to taxes in
maximizing
the
expected
net
benefit of
control,
and if
they
are allocated rather than
auctioned,
they
involve a
lump-sum
transfer to
industry,
of a kind
comparable
to
giving
away
rather than
selling
state-owned assets
upon privatization.
If
instead,
taxes are used
(or
licenses are
sold)
then the
government
can
reduce other taxes
by
the amount of extra
revenue,
and so reduce
deadweight
losses elsewhere in the
economy.
Put another
way,
most
taxes raise revenue at the cost of some
inefficiency,
but corrective taxes
raise revenue while
improving efficiency.
The latter are
highly
attrac-
tive sources of
revenue,
not to be
lightly forgone.
There is a further
limitation which has not
yet
been
mentioned, arising
because of the
nature of the abatement
technology.
Flue
gas desulphurization
or
denitrification
requires heavy
fixed investment with a lifetime
equal
to
Acid rain 331
that of the
plant-perhaps up
to 40
years.
Its cost-effectiveness will
depend
on the costs of emission in the
future,
and these
may
be harder
to
predict
if
they
are determined
by
a
spot
market in licenses at each
future
date,
than if
they
are determined
by
a tax which is announced
and
expected
to remain
relatively
constant.'4
This
objection
can be
overcome
by issuing plants
with an estimated
(but non-negotiable)
lifetime
supply
of
licenses,
so that the
plant
owner would
expect
neither
to be a net seller nor net
buyer
of licenses for the
expected
lifetime of
his
plant.
The
difficulty
with this solution is that licenses for new
plant
would have to come from the
liquidation
of allocations from
existing
plants, though
one should not
exaggerate
this
difficulty.
Old
plants
will
have to be
retired,
and would at the
predetermined appropriate
date
release licenses for reallocation.
Licenses work well for
large stationary
sources which are the main
source of
SO2,
but are not
immediately applicable
to small mobile
sources,
like vehicles which are
increasingly responsible
for
NOx.
Licenses could be issued or auctioned to car manufacturers who would
fulfil their
quota,
T tonnes
NO2
equivalent per annum, by selling
n
vehicles each of which emits no more than t tonnes
per
annum of
normal
driving,
where nt
<
T.
Subsequent
vehicle
testing
would ensure
that emissions were
kept
down to t.
Given the
advantages
and
disadvantages
of taxes and
licenses,
is there
some
compromise system
which
might
do better than either
pure
instrument? The
anti-competitive problem
can be met
by
the
govern-
ment or its
agency being willing
to
buy
or sell licenses at a
predetermined
price.
This would
require
an international
agreement
about the form
in which abatement would be
encouraged.
If such
agreements
insist on
quantity controls,
then the
agency
would find it difficult to hold the
price
of licenses down to a
satisfactory
level needed to
prevent
the
incentive to raise rivals' costs. If the international
agreements merely
defined the status
quo
for
bargaining,
then
recipient
countries could
establish a market
price
for
licenses,
and
might,
if
they
overcame their
coordination
problems,
be able to act as a
quasi competitive
force.15
This,
quite apart
from the increased
efficiency
of such a
system, provides
another
argument
for
attempting
to
negotiate
such treaties.
The main
arguments against using pollution
taxes are that
recipient
countries
may
fear that it has little effect on
pollution levels,
while the
polluting companies
see the tax as
harming
their
profits.
The first
worry
I I
14
Indexing
the tax to
per capita
income in
recipient
countries to allow for
increasing damage
costs would not
substantially
reduce
predictability.
15
If
damage per
tonne is
relatively independent
of both
pollution
levels and the
year
in which
it
occurs, recipient
countries could
arbitrage
the
price
over time. Relative
stability
of the
price
would reduce the
ability
of dominant
polluters
to
manipulate
the
price
of licenses.
332 David
Newbery
can be
greatly
alleviated
by combining
the tax with emission standards
on new
plant, designed
to be
cost-justified
at the tax rate.
(It may
be
enough
to set limits which firms would choose to better
given
the
tax,
but which are sufficient to
allay
doubts on the
part
of
recipients.)
The
second
problem
is
primarily
one of
dealing
with firms with
existing
inappropriate plant.
It
may
be sufficient to draw
up
a reference emission
plan
for each
large plant, specifying
the maximum allowable emissions
in each of the next 20
years (perhaps falling by
5%
of the initial value
each
year,
down to
zero).
The
plant
would be entitled either to licenses
or a tax rebate to this
amount,
but would have to
pay
for emissions
above this level.
3.2.4. Licenses and
charges.
Hahn
(1989) provides
a useful
survey
of the
experience
in
dealing
with environmental
problems
in four countries.
The instrument most economists find attractive is the marketable
per-
mit,
and he describes two
important applications
in the US. In 1981
Wisconsin allowed firms to trade the
rights
to
pollute
the Fox
River,
but
resulting cost-savings
were minimal. Hahn finds this
unsurprising,
as the main sources were
regulated municipal
utilities and
oligopolistic
paper
and
pulp
manufacturers. The
arguments
above
suggest
that in
both cases the incentives to trade to achieve
efficiency
will be
muted,
and it is
interesting
to have this
empirical
confirmation.
A much more
important experiment
was the US Environmental
Protection
Agency's
emissions
trading policy,
with its various com-
ponents allowing
firms to meet firm level or area level standards
by
either internal
reorganization
of their
pollution
activities or
by
external
trading
or
bargaining.
The
resulting cost-savings
have been
substantial,
and Hahn cites estimates of
cost-savings
of between
$525
mn. and
$12 bn.,
the wide
range reflecting very incomplete monitoring
of the
cost-savings
achieved within firms. Most of the
gains
came from internal
trading,
and
pressures
from environmental
groups appears
to have led
regulators
to
downplay
the
property rights
nature of
permits,
thus
undermining
their
marketability.
Hahn concludes that 'emissions trad-
ing
is best viewed as an incremental
departure
from the
existing
approach. Property rights
were
grandfathered.' (Hahn, 1989,
p. 101.)
The most successful
example
of
permit trading
was lead
trading
in
gasoline,
where refineries could
buy
or sell the
right
to lead levels in
gasoline,
based on an
existing
standard
per gallon.
About 15% of total
lead
rights
used were
traded,
with an active
spot
market and an intertem-
poral
market via
'banking'.
The
systems appears
to have worked well
because it is
easy
to monitor and there was
widespread agreement
about
the
objectives
of the
programme,
which were to
gradually phase
out
lead in
gasoline. Perhaps
the most difficult
aspect
of environmental
Acid rain 333
regulation
is this
problem
of
reaching
a well-defined consensus on what
to do - much of the environmental
opposition
arises because of distrust
as to the
objectives
of the various
parties.
If marketable
permits
have been rather
disappointing,
there
appears
to have been
reasonably widespread acceptance
in
Europe
at least of
charges
for
pollution (primarily
for water borne and solid
pollutants).
The revenue from these
charges
is
normally
earmarked for environ-
mental
cleanup,
so the link between the
charge
and the
resulting cleanup
is seen to be
fair, given
that it would have
presumably
cost the
polluter
more to clean
up
the
pollution
itself.
3.3. A
possible strategy
To
date, inter-government negotiations
have been
guided by
the
prin-
ciples
of uniform reduction in emissions from a benchmark level
(usually
1980 actual
emissions)
for
large
combustion
plants,
or towards uniform
standards for mobile sources
(i.e. cars).
The attraction of this is that it
appears
to
impose equal
costs on the
participants,
so that it
gives
no
commercial
advantage
or
disadvantage
to
producers
in each
country.
In
fact,
of
course,
countries which have become
heavily dependent
on
nuclear
power (France), hydroelectricity (Scandinavia),
or
gas (the
Netherlands),
can achieve these emission reductions at lower costs than
those,
like the UK and
Germany,
still
heavily dependent
on coal. There
are two serious
problems
with this
approach.
The first is
political,
and
may
be sufficient to derail the
negotiations-some
countries are net
losers from such
negotiations.
The second
objection
is economic- the
reduction in
aggregate pollution damage
is done at
higher
than least
cost. Both
problems
can be solved
by
the natural solution which allows
beneficial and efficient
bargaining.
This would involve first
agreeing
a
benchmark
trajectory
of allowable emissions
(not necessarily equal
for
all
countries),
and then
facilitating bargaining
over deviations from this
level. This
might
best be done
by
first
estimating
the
marginal damage
costs
(or willingness
to
pay
for
abatement,
if
higher) by
EMEP
cell,
then
calculating
the
appropriate
cost
sharing
formula for the
group
of coun-
tries affected
by polluter j.
Thus for
country
i its share
ai
of the
payments
to
polluter j
would be
ai =
MDitij/
MDktk
k
The
recipients
would then
appoint
a
negotiator
with
powers
to
levy
charges
on
recipients proportional
to these cost
fractions,
up
to the
total
damage
level. The
negotiator
could then bribe the
polluting
country
to make additional reductions.
334 David
Newbery
The main
problem
lies in
decentralizing
this within the
polluting
country.
One solution would be to
compute
cost-effective emission levels
for new
plants (i.e.
the level at which the
marginal
cost of further
abatement is
equal
to the
marginal damage)
and set this
(or
a somewhat
more lenient
standard)
as the benchmark for new
plant.
Deviations
from this would then be taxed
(if excessive)
or rebated
(if lower).
Old
plant
could be
similarly
dealt
with,
at least for the
large
combustion
plants,
which are not too numerous in
any country. Again,
the first
step
is to determine the
benchmark,
presumably starting
near current
emission
levels,
but
falling
to zero at the estimated economic date of
plant
retirement. Where
retro-fitting
is
evidently cost-effective,
then
the benchmark
might
be the retro-fitted emission level. The emission
tax would be uniform over all
plants,
new and
old,
but would
vary
with
location.
The current tax rate would be the amount offered
by
the
recipient
negotiator.
The main
problem
is that FGD
requires
a
large
fixed invest-
ment to amortize over
up
to 40
years,
and thus
requires
the investor
to calculate the future benefits
(i.e.
taxes
avoided)
of installation. If
these future benefits are
uncertain,
then abatement
may
be deterred.
It would
pay
the
recipient negotiator
to offer formal
contracts,
possibly
indexed to
recipient GNP,
in order to reduce
uncertainty
and thus
induce
greater
abatement for the same cost. This would not
completely
solve the
problem,
at least in the
UK,
for the
following
reason. The
argument
advanced above
suggested
that it
might
be in the interests
of the
duopoly electricity generators
to face
competition
from a
high
cost
fringe
of
producers.
It
might
therefore
pay
National Power and
Powergen
to sell off their least
efficient,
most
polluting plant
to
indepen-
dents. The
buyers
would be uncertain of the
likely
future demand for
their
power (assuming they
are
unlikely
to secure
long-term supply
contracts with the Discos for more than a small fraction of their
output).
Even if
they
knew the future
payment per
tonne
SO2 reduced,
they
would be uncertain of their
likely
future
electricity
(and
hence
SO2)
production. They
are
unlikely
to install
FGD,
and their
operating
costs
will thus be raised
(by
the emissions
tax).
This in turn will raise the
equilibrium price
of
electricity,
and increase the
profits
of National
Power and
Powergen.
In
short,
the
asymmetry
in abatement costs for
new base load
plant
and old coal-fired
plant
makes the case for
raising
rival's costs
stronger
than that
given
in the
Appendix.
One should not
exaggerate
this risk.
Encouraging
divestiture
by
the
duopoly generators
will at least increase the number of
competitors,
and should make
entry-deterrent
collusion more
difficult,
for the future
benefit of consumers. It
may
also be that
retro-fitting
is not cost-effective
for the kind of
plant likely
to be
sold,
in which case the
problem
does
Acid rain 335
not arise. Nor does it seem
likely
that emissions standards and taxes
would discriminate
against
new
entrants,
because the best
practice
generating technique
for new entrants
appears
to be
high efficiency gas
turbine or
gas-based
combined heat and
power systems,
of modest
capacity (130-300 MW).
Emissions from such
systems
are
naturally
low,
giving
them an additional
advantage
over the
existing large
coal-fired
conventional
power
stations.
If this seems
utopian,
it nevertheless
appears likely
that some inter-
national
agreement
on emissions will be
reached,
and it is in
any
case
in each
country's
own interest to reduce
emissions,
as such a
large
fraction is
deposited
within its own borders. This
being
the
case,
coun-
tries like the UK which
have,
or will
have,
a
privately
owned
electricity
supply industry
will still need a decentralized
system
of abatement.
Standards,
appropriately
chosen to be
cost-effective, may
be
satisfactory
for new sources
(and
especially
for small
sources,
like vehicles where
the cost of
individually
monitored
charges
is
likely
to make them
uneconomic). They
are not so suitable for
existing
sources where some
abatement is cost-effective. Here taxes
(or charges)
have obvious advan-
tages, though bargaining
between the relevant
government agency (Her
Majesty's Inspectorate
of
Pollution)
and the individual firm
may
be a
satisfactory
alternative. The attraction of an
explicit
tax or
charge
is
two-fold-it
signals
to
recipient
countries the
advantage
of further
negotiations (with side-payments)
and it
signals
the
advantage
of
developing
more cost-effective abatement
strategies.
In their
absence,
the
generators
themselves have no such
inducement,
as
they
risk
having
the new
technology
mandated in more
stringent
standards
(now
cost-
effective).
4.
Summary
and conclusions
In
many ways,
the debate on acid rain is
quite encouraging
from an
economists'
perspective.
EMEP has
quantified pollution
flows and
pro-
vided the raw data for bilateral
bargaining.
The least
satisfactory
ele-
ment so far is that of
quantifying
the
damage
costs
(rather
than the
levels of
deposition).
Much commendable effort has been directed to
identifying
the
ecological pathways by
which acid rain harms the
ecological environment,
but
insignificant
effort to
quantifying
the cost
of this harm. It
appears
that
property damage
is more
important
than
ecological damage,
and that forest and
crop damage
is far
greater
than
lake and river
acidification,
yet very
little recent work is directed to
quantifying property damage.
No doubt this reflects the emotional level
at which most environmental issues are
discussed,
but one
might hope
that the best
way
to lower the emotional
temperature
is to increase the
factual and economic content of such debates.
336 David
Newbery
The main
findings reported
here are that it is
likely
to be
excessively
costly
to aim at a uniform reduction in emissions in all countries. Since
some countries will
gain
while others lose from such
proposals,
it is
likely
to be
politically
difficult or
costly
to reach
agreement. Instead,
it
seems more sensible to aim at
reducing
emissions where the benefits
of reduced
damage
exceed the
marginal
costs of
abatement,
and to aim
at methods of
coordinating payments
for abatement to overcome the
'free-rider'
problem.
The other
finding,
less secure
given
the
present
state of scientific
knowledge,
is that
reducing NOx appears
less
urgent
and less attractive
than
reducing
SO2.
Certainly,
the cost of
reducing
automobile emissions
is
high,
the health
gains small,
and the
ecological
benefits uncertain.
On the other
hand,
it
may
be
cheap
to reduce
NOx
emissions from
stationary sources, using
better burner
designs
and chemical additives.
If one
priority
is to be
singled
out,
it is that of
reducing
the
appalling
pollution
in Eastern
Europe
in order to secure
primarily
health benefits.
Given the
inefficiency
of
energy
use in these
countries,
it is
likely
that
substantial reductions can be achieved at
relatively
low cost and
high
benefit,
in
many
cases
just by closing
down
uprofitable enterprises,
and
perhaps moving
towards a more
integrated energy market, relying
more on trade and less on
self-sufficiency.'6
Discussion
Horst Siebert
University
of Kiel
This
paper provides
valuable evidence on the
magnitudes
of inter-
national
spillovers
in airborne
pollution.
The main contention of the
paper
is that a uniform reduction in emissions from
stationary
sources
is an inefficient solution to the
problem.
This
(a
standard result in
environmental
economics)
is for two reasons:
marginal
abatement costs
may
differ between
countries,
and a unit of
pollutant
will cause different
levels of
damage according
to where it is
deposited.
The Polluter
Pays Principle
is
impossible
to
apply
in international
contexts because there exists no
supra-national authority
to attribute
environmental costs to the
polluter
via taxation or
regulation.
Inter-
national
spillovers
constitute a
game-theoretic problem
in which the
non-cooperative
solution is
inefficient,
because each
country
does not
take into account the effects of its actions on the welfare of other counties
when it seeks to minimize the costs of
pollution
net of abatement costs.
S H 1 f a d o e a
16
See
Hughes (1990)
for a discussion of
energy
and environmental issues in Eastern
Europe.
337
Acid rain
In a
cooperative
solution the net costs are minimized for all countries
together.
But this
implies
that there
may
be countries that are worse
off in the
cooperative
than the
non-cooperative
solution,
in the absence
of side
payments
(an
upwind country
such as the UK is an obvious
example).
The need for
side-payments
means therefore that the Victim
Pays Principle
is more
appropriate
for
solving
international
spillover
problems.
However,
there
may
be solutions intermediate between the
coopera-
tive and the
fully non-cooperative
ones. There
may
be coalitions of a
subset of countries who would benefit from
cooperation
to reduce
emissions. This is
particularly likely
if countries interact
repeatedly
and
therefore have
reputations
to
preserve.
Furthermore, cooperation
in
the environmental field
may
be offered
by
some countries in
exchange
for
cooperation
in other fields
(those
of trade or
political relationships,
for
instance). According
to Maler
(1989)
the US
agreed
to build a
desalination
plant
on the Colorado River in order to
improve
its
political
relations with
Mexico,
for instance.
How can a
cooperative agreement
be
implemented? Newbery
dis-
cusses a number of
systems,
none of them without
problems.
As an
alternative to those he describes I should like to
suggest
the use of
international diffusion norms. A diffusion norm
specifies
the ambient
quality
of an environmental medium
(air
or
water)
when it crosses an
international
boundary,
and
thereby
limits the allowable level of inter-
national
spillover.
Diffusion norms have been used in national water
quality manage-
ment. In
Germany,
for
instance,
the ambient
quality
of a
tributary
to
the Rhine
(the Emscher)
is
specified
where it enters the main river. It
would be
easy
to
apply
the same
principles
to the ambient water
quality
of rivers such as the Rhine or the Thames where
they
enter the North
Sea. It should not be much more difficult to establish
monitoring
stations
to measure airborne
spillovers.
The attraction of international diffusion
norms is that
they
allow environmental
policy
to be decentralized to
individual
countries,
which are able
(it
is
assumed)
to enforce
coopera-
tive solutions in a domestic
setting.
It can be left to each
country
to
decide how it wishes to secure a
specific
environmental
quality
at its
border.
One
way
to think of environmental issues in the context of the
international division of labour is to consider individual countries as
being differentially
endowed with environmental factors as well as
different national
preferences
for environmental
quality.
In the absence
of international
spillovers,
national differences in environmental
endowments are
just
like differences in
any
other endowment. The
price
of environmental
protection (such
as a tax on
emissions)
is the
338 David
Newbery
price
for an immobile
factor,
and it is
quite
normal for the
prices
for
immobile factors to differ
among
countries.
Unfortunately,
this attrac-
tively simple conception
is
complicated by spillovers.
But international
diffusion
norms, by limiting spillovers,
allow us to decentralize environ-
mental
policy
and reduce the need for a new international
bureaucracy.
John Vickers
Nuffield
College,
Oxford
The economics and
politics
of the acid rain
problem
in
Europe
are
fascinating
for a number of reasons:
(1)
The
damage
caused
depends
not
only
on how much but also on
where the
pollutants
are emitted.
(2) Deposits
from
any given
source
typically
occur at
many
locations
and
change unpredictably
with the weather.
(3) Intertemporal
considerations matter because acid rain is a stock
pollutant.
Abatement
technologies may
also have
long
lifetimes.
(4)
The chief
polluter-the electricity supply industry
- is not
per-
fectly competitive.
(5)
Policies have
major
distributional
consequences
both between and
within countries
(for
the
latter,
consider miners of
high sulphur
coal).
(6)
A two-tier
policy
is
needed-agreement
between countries and
implementation
within countries.
David
Newbery provides
an informative account of the
problem
and
a
persuasive argument against
uniform emission reductions as a sol-
ution. But it is easier to
say
what is not the solution than what is.
Newbery
considers the relative merits of emission
standards,
taxes and tradeable
permits.
The theorist is
initially
attracted to
permits
because of their
informational
economy
under
competitive
conditions. But
Newbery
points
out four drawbacks:
(a) They
are vulnerable to
anti-competitive strategic
behaviour
by
firms
preempting
licenses to raise rivals' costs.
(b)
Under
uncertainty they
are inferior to taxes if the
marginal
cost
of abatement schedule is
steeper
than the
damage
schedule,
for
certainty
over control costs is then more
important
that
certainty
over emission levels.
(c) They
raise less revenue than
taxes,
unless sold.
(d) They
make
intertemporal
calculations
difficult,
because future
permit prices
tend to be harder to
predict
than future taxes.
The seriousness of
(a) depends
on how
competitive
the
permit
markets
are. This in turn
depends
on the
geographical scope
of the
market,
a
Acid rain 339
factor considered further below. Point
(b)
is an
empirical
matter:
Newbery argues
that the
age profile
of
plants suggests
the abatement
schedule is indeed
steeper,
which is
plausible
but
open
to debate. A
response
to
(c)
is to
say
that
permits
should indeed be
sold,
but that
vested interests will resist. Another
difficulty, affecting (d),
concerns
time-consistency: having
sold a
given
number of
permits,
a
government
might
be
tempted
to sell some more for revenue
reasons,
and
might
be unable to commit not to do so.
There is
yet
another
problem,
which is not
unique
to
permits: (e)
Location
matters,
since acid rain is a non-uniform
pollutant.
Permits
can in
principle
be individuated
by
location. The
theoretically
ideal
way
to do this is
by
location of
deposition,
but volatile weather
patterns
make this
practically impossible.
A more realistic
suggestion
is to
individuate
permits by
location of emission. But there is then the
difficulty
of
determining
the relative
price
of
permits
for emission at
different locations.
If the relative
price
is determined in the
market,
with
separate
markets
for each
location,
each market
may
be
seriously uncompetitive.
The
more
competitive
the
permit
markets,
the less
sensitivity
to location.
Alternatively,
there could be a
single type
of
permit (for
which the
market would be
large
and
reasonably competitive),
and the number
of
permits required
to emit a ton of
sulphur
would
vary by
location.
The determination of this number would have to be
administrative,
since countries downwind of the most
damaging
locations would not
be able to reduce the emissions from a
specific
location
merely by trading
in the market. Even
Newbery's proposed compromise
-a benchmark
trajectory
of
emissions,
with
bargaining
over deviations-would face
this
problem.
Because of the
multiplicity
of
deposition
locations from
any
one
source,
countries
wishing
to achieve a reduction from that
source would have to band
together,
which would
require
them to share
out the costs of
doing
so on the basis of relative
damage
assessments.
In
any system
sensitive to
location,
estimation of this latter element
cannot be avoided.
Finally,
it is
important
to be realistic about some of the
political
obstacles involved. Even in the
US,
successful
examples
of tradeable
permits
have not been common. And it is easier for the Federal Govern-
ment to curb emissions from Midwestern
power plants
than for Brussels
(or anyone else)
to do the same for
British,
let alone East German
plants.
General discussion
Many panellists
were concerned about the
reliability
of the estimates
of
marginal damage
and
marginal
abatement costs. Richard Baldwin
340 David
Newbery
pointed
out that GDP-based measures of
damage ignored
the value of
leisure: this was
particularly important
in
estimating
the
damage
to
forests. David
Newbery
said that he
suspected
the leisure costs were
highly
correlated with GDP.
John
Black was concerned about
equity:
it seemed unfair to
propose
that
pollution
inflicted on the
poorer
citizens
of Eastern
Europe
was more
acceptable
because of their
poverty.
David
Newbery replied
that his
paper
was concerned
only
about
efficiency
aspects
of the
problem,
and GDP
per
head would be what determined
individual countries'
willingness
to
pay
to reduce
sulphur
dioxide
emissions.
Richard
Jackman thought
that abatement costs in Eastern
Europe
might
be overestimated: much
pollution
had been a
by-product
of
economic
activity
that would not be
profitable
in a more market-based
economy
and could therefore be
expected
to decline in
any
case. He
suspected damage
costs
might
be
highly
non-linear,
particularly
those
that concerned
damage
to health. And he also
thought
it
important
to
consider
intertemporal aspects
of the
problem,
since
many
Eastern
European
countries
might expect
to
grow
much faster in the
future;
in
addition,
because acid rain was a stock
pollutant
reductions in
emissions
might
take a
long
time to affect
damage
levels. David
Newbery
agreed,
and
emphasized
that much of the
acidity
in lakes
today
was
due to emissions of
pollutants
from as
long ago
as the industrial revol-
ution.
Fortunately, liming
of lakes could
speed up
the reversal of the
process,
and
atmospheric pollution (which
was more
damaging
to
property)
was much more
closely
related to current emissions.
There was some discussion of the
magnitudes
of the
aggregate
costs
involved.
Jean-Charles
Rochet said that
although
the
paper
had shown
that a uniform emission reduction was not
optimal,
it
might
nevertheless
not be far from the
optimum
in overall welfare terms. And a uniform
solution would
certainly
be easier to make
politically acceptable.
David
Newbery
said it was not
easy
to estimate the costs of a
sub-optimal
as
opposed
to an
optimal
solution to the
problem,
but the fact that the
recommended reductions in Table 7 varied so
widely suggested
that
the cost of a
sub-optimal
solution was
likely
to be substantial. He was
also not convinced of the
insuperability
of the
political
obstacles to
solutions
involving bargaining.
What was
necessary
was for environ-
mental
protection agencies
to
emphasize
that the control of
pollution
was a
good
like
any other,
and that it made sense to
produce
it
efficiently.
Once this was done it should be
possible
to ensure a stable framework
for
trading permits; past
difficulties in the US in
ensuring adequately
thick markets had been exacerbated
by uncertainty
over their continued
operation
in the future.
Acid rain 341
Table A.1.
Origins
of
sulphur deposition
in
Europe (thousands
of tonnes of
sulphur per year)
Emitters
Receivers AL AT BE BG CS DK Fl FR DD DE GR HU IS IE IT LU NL NO PL PT RO ES SE CH TR SU GB YU NA UI Total
Albania AL
Austria AT
Belgium
BE
Bulgaria
BG
Czechoslovakia CS
Denmark DK
Finland FI
France FR
GDR DD
FRG DE
Greece GR
Hungary
HU
Iceland IS
Ireland IE
Italy
IT
Luxembourg
LU
Netherland NL
Norway
NO
Poland PL
Portugal
PT
Romania RO
Spain
ES
Sweden SE
Switzerland CH
Turkey
TR
USSR* SU
Britain GB
Yugoslavia
YU
N. Africa** NA
Rounding
errors
Total
Production 1987
5 0 0 1 1 0 0 0 1 1 1
0 18 2 0 27 0 0 12 24 18 0
0 0 50 0 2 0 0 19 7 15 0
0 0 0 152 4 0 0 0 3 1 3
0 4 4 0 385 0 0 11 128 28 0
0 0 1 0 2 31 0 1 12 7 0
0 0 0 0 4 1 48 0 8 3 0
0 0 23 0 19 1 0 332 41 40 0
0 0 7 0 84 2 0 14 725 61 0
0 1 29 0 47 3 0 69 163 330 0
0 0 0 17 2 0 0 0 2 0 45
0 3 1 1 31 0 0 3 16 6 0
00 0 0 0 0 0 0 1 0 0
0 0 1 0 0 0 0 2 2 1 0
0 1 2 0 13 0 0 21 15 8 0
0 0 0 0 0 0 0 2 0 1 0
0 0 15 0 2 0 0 11 8 35 0
0 0 1 0 5 6 2 2 15 6 0
0 3 6 1 145 5 0 15 310 47 0
00 0 0 0 0 0 0 0 0 0
0 1 0 22 28 0 0 3 18 5 2
0 0 1 0 2 0 0 11 5 3 0
0 0 2 0 12 13 8 3 33 12 0
0 0 1 0 2 0 0 13 4 4 0
0 0 0 13 3 0 0 1 4 1 6
0 3 5 17 107 9 22 10 167 36 3
0 0 5 0 5 0 0 14 15 11 0
1 3 1 18 22 0 0 8 18 8 4
3 5 39 37 105 43 18 136 253 131 26
3 5 6 4 5 7 3 8 7 4 4
12 47 202 283 1,064 121 101 721 2,005 823 94
25 75 244 570 1,450 155 162 923 2,500 1,022 180
1 00 4 0 0 0 1 0 0 0 0 0 0 0 0 0 0 8 30
12 0 0 30 0 1 0 15 0 0 2 0 0 0 0 4 8 0 27 207
0 00 0 0 4 0 2 0 0 1 0 0 0 0 11 0 0 8 121
10 0 0 5 0 0 0 5 0 4 0 0 0 1 5 0 14 0 22 235
45 0 0 10 0 1 0 95 0 0 1 0 0 0 2 7 9 0 28 765
0 00 0 0 1 0 4 0 0 0 0 0 0 1 7 0 0 9 83
2 00 0 0 0 0 12 0 0 0 3 0 0 57 3 1 0 62 210
5 0 1 21 0 5 0 15 2 0 65 0 1 0 0 43 3 1 139 760
2 00 2 0 4 0 32 0 0 1 0 0 0 1 15 1 0 24 979
3 0 0 13 1 14 0 23 0 0 6 0 1 0 1 45 1 0 64 821
3 00 5 0 0 0 2 0 0 0 0 0 1 2 0 6 0 28 119
190 0 0 12 0 0 0 25 0 1 0 0 0 0 1 1 23 0 18 337
0 0 0 0? 0 0 0 0 0 0 0 0 0 0 0 3 0 0 15 21
0 0 21 0 0 0 0 1 0 0 0 0 0 0 0 17 0 0 19 68
11 0 0 353 0 0 0 14 0 0 10 0 2 0 1 4 16 3 83 562
0 00 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 7
0 0 0 0 0 32 0 2 0 0 1 0 0 0 0 20 0 0 10 139
1 0 0 1 0 1 13 11 0 0 0 5 0 0 10 19 0 0 91 194
40 0 0 10 0 ,4 0 790 0 1 1 1 0 0 18 15 11 0 64 1,492
0 00 0 0 0 0 0 35 0 19 0 0 0 0 0 0 0 26 83
61 0 0 13 0 0 0 33 0 34 1 0 0 1 17 1 39 0 43 330
2 00 2 0 0 0 3 15 0 523 0 0 0 0 6 1 1 97 674
3 0 0 1 0 2 4 33 0 0 0 37 0 0 23 13 1 0 103 307
0 00 14 0 0 0 1 0 0 3 0 8 0 0 3 0 0 14 70
5 00 4 0 0 0 3 0 1 0 0 0 61 12 0 5 1 86 210
84 0 0 13 0 3 1 337 0 12 1 7 0 7 2,204 16 24 0 491 3,584
0 0 6 1 0 3 0 3 0 0 2 0 0 0 1 571 0 0 60 702
47 0 0 61 0 0 0 24 0 1 4 0 0 0 1 32 192 1 74 497
64 0 17 182 0 32 7 194 13 6 210 21 1 23 196 437 62 11 810 3,087
3 0 6 2 3 8 3 5 2 5 5 5 4 1 5 -22 7 3 180 31
594 0 51 759 5 115 28 1,685 67 65 856 79 17 95 2,558 1,271 424 21 2,532 16,695
710 3 84 1,252 6 141 50 2,270 116 100 1,581 116 31 177 5,100 1,840 588 0 21,471
Source: Acid
Magazine
No.
8, Sept. 189, p.
8.
Notes: As for Table 1.
Appendix.
Data details and formal models
1. Data details
Table A.1
provides
the
comprehensive
data on the
origins
and
deposi-
tions of
sulphur
in
Europe,
from which the
summary
tables of the text
are calculated.
2. The choice between taxes and
quotas
under
uncertainty
Suppose
that the abatement schedule
gives
the
marginal
cost of abate-
ment
p
in terms of the level of
abatement, x,
as
p
= a + bOx +
e,
EO
=
1,
Ee
=
O,
E0E
=
O
where 0
represents
the
uncertainty
about the
slope parameter
b and e
represents uncertainty
about the level of initial
cleanup
costs,
assumed
independent.
If w is the
willingness
to
pay
for
cleanup,
and X is the
current level of
emissions,
then
suppose
that
w
=
c + r + d
(X-x),
E 1 =
1, E7
=
O, ErO
=
0
Again,
X
represents uncertainty
about the
sope,
7
about the level when
there is no
pollution (i.e.
x =
X).
Assume that all four random variables
are
independently
distributed.
The net social
surplus (consumer surplus plus profit)
when the level
of abatement is x is
given by
= (w
-
p) dx
=
S(x)
o
If the
government
must choose the level of
quota, x,
before
knowing
the
parameters
of the
system,
then x is chosen to max
ES(x).
The
solution is
c+dX-a
(c+dX-a)2
b+d
'
2(b+
d)
If the
government
must choose a tax rate
t,
then x =
x(t), given by
t-a-e dx 1
bO' dt
bO
The value of t is now chosen to max
ES[x(t)],
and the result is
(c
+
dX-a)2(E 1/)2
(c
+
dX-a)2(1
+ 20o2)
ES=
2[bE(1/8)+ dE(8-2)] 2[b(1
+
o2)+
d(1
+
3a2)]
where r2=
E(O- 1)2
is the coefficient of variation
squared
of the
Acid rain 343
abatement
slope.
Taxes will be
superior
to
quotas
if
ES(t)> ES,
i.e. if
b[(Eof)2-EqI]>d[E 2-(E4l)2],
=
1/0
A sufficient condition for this to hold is that a is small and b >
d,
i.e. if
the
slope
of the abatement cost schedule is
steeper
than the
slope
of
the
damage
schedule.
3. A
duopoly
model
Consider the
following very simplified
model of a
duopoly electricity
supply industry.
Demand is
linear,
and the market
clearing price
is
p
= a -
Q
when
aggregate supply
is
Q.
The
technology
is one of constant
unit
operating
costs
(set equal
to zero without loss of
generality) ignoring
pollution
abatement. The amount of
pollution
released when
output
is
q
and abatement
equipment
h is installed is
q/h.
The unit cost of
abatement
equipment
is
r,
and the
generator
is allowed to release x
units of
pollution per year.
Its
profit
is then
pq
- rh
=
(p
-
r/x)q.
Each
generator
behaves as a Cournot
duopolist
in the
output market,
given
its entitlement to release
pollutant,
and chooses
q
to maximize
profit,
assuming
the other
generator's output
is
given.
If the other
generator
can release
y
units of
pollution,
then
1
(
2 11
)
and the market
clearing price
is
p=-
(
a+r -+)
3 , x
providing
in both cases that the other
generator produces
a
positive
output.
If
not,
then the
generator
is assumed to behave as an uncon-
strained
monopolist.
If the total
pollution
level is fixed as 1
unit,
so
that
y
= 1 -
x,
then individual
profit, aggregate profit,
consumer
surplus,
and net social
surplus (the
sum of consumer
surplus
and
aggregate
profit)
are all functions of x. At the
symmetric duopoly equilibrium
x
=0.5, locally
net social
surplus
is
maximized,
while net
aggregate
profits
are minimized
given
the chosen
parameter
values
(a
=
10,
r=
1.5). Figure
A.1 shows the
resulting plot.
There is a
discontinuity
as
the
duopoly collapses
into a
monopoly,
and it is assumed that the two
firms are not allowed
effectively
to combine to form a
monopoly,
and
the
graph
is terminated at the
point
at which
monopoly
could occur.
It is an
interesting
observation that if the total licenses were to be
allocated to the
surviving firm,
then the net social
surplus may
increase
344 David
Newbery
Acid rain 345
25 -
Social
surplus
20 -
Un
a
15-
),
Cons
surplus
Total
profits
10 - .
-
.......
......
. -
O_:
......-*****""""""?'
Profits 2
5-
Profits 1
0 '
I
0.5 0.45 0.4 0.35 0.3
Fraction of licenses in firm 1
Figure
A.1. Emissions license reallocations
as the combined
inefficiency
of the
duopoly
and the misallocation of
pollution
licenses is
replaced by
the
single inefficiency
of a
monopoly.
The fact that
aggregate profits
increases as the
duopoly
moves
away
from the social
optimum
towards an
asymmetric equilibrium
means
that there are
profitable exchanges
of
pollution
licenses for cash between
the two
duopolists
which leave each better off than in the
symmetric
equilibrium.
It will
typically pay
one firm to
buy
all the licenses from
the other firm in order to become an unconstrained
monopolist.
If this
is not allowed
by competition policy,
then the firms will aim at the
maximum allowable
degree
of
asymmetry.
References
Alfsen,
K.
H.,
D. A. Hanson and S. Glomsr0d
(1986).
'Direct and indirect effects of
reducing SO2
emissions:
experimental
calculations on the MSG-4E model', Central Bureau of Statistics
Discussion
Paper 20,
Oslo.
Amann,
M. and G. Kornai
(1987).
'Cost Functions for
controlling
SO2
emissions in
Europe',
Working paper WP-87-065,
International Institute for
Applied Systems Analysis, Laxenburg,
Austria.
Battarbee,
R. W. and 15
colleagues (1988).
'Lake acidification in the United
Kingdom
1800-1986',
Report
of the
Palaeoecological
Research
Unit, University College London, HMSO,
London.
Berkhout, F.,
S. Boehmer-Christiansen and
J.
Skea
(1989). 'Deposits
and
repositories: electricity
wastes in the UK and West
Germany', Energy Policy.
Brackley,
P.
(1987).
Acid
deposition
and vehicle emissions:
European
environmental
pressures
on
Britain,
Aldershot: Gower.
Britt,
D.
(1986).
'The renovation of acid lakes and
streams', chapter
2 of Acid rain Control II: The
promise of
new
technology,
D. S. Gilleland and
J.
H. Swisher
(eds.),
Southern Univ.
Press,
Carbondale.
346 David
Newbery
Coase,
R.
(1960).
'The
problem
of social cost'.
Journal of
Law and Economics.
Office
(1986). Curbing
acid rain:
cost, budget
and coal-market
effects,
US
Government
Printing Office, Washington
DC.
Crandall,
R.
W.,
H. K.
Gruenspecht,
T. E. Keeler and L. B. Lave
(1986). Regulating
the
Automobile,
Brookings, Washington
DC.
Department
of the Environment
(1988).
Our Common Future: A
perspective by
the UK on the
Report
of
the World Commission on Environment and
Development.
Derwent,
R. G.
(1988).
'A better
way
to control
pollution',
Nature.
(1990). Optimal strategies for sulphur
emission control in
Europe,
Harwell
Laboratory.
Dowlatabadi,
H. and W.
Harrington (1989).
'Policies for the
mitigation
of acid rain',
Energy Policy.
Dudley, N.,
M. Barrett and D. Baldock
(1985).
The Acid Rain
Controversy,
Earth Resources
Research,
London.
(1983).
Acid rain: a review
of
the
phenomenon
in the EEC and
Europe, prepared
for the Commission of the
European Communities,
Graham and
Trotman,
London.
HMSO
(1984). Royal
Commission on Environmental Pollution - Tenth
Report,
Cmnd
9149, London,
HMSO.
Hahn,
Robert W.
(1989).
'Economic
Prescriptions
for Environmental Problems: Ho; the Patient
Followed the Doctor's
Orders', Journal of
Economic
Perspectives.
Hughes,
G. A.
(1990). 'Energy policy
and the environment in
Poland', European Economy.
IME
(1987).
'Vehicle emissions and their
impact
of
European
Air
Quality',
Institution of
Mechanical
Engineers
Conference Publications 1987-88.
Jeffrey (1988).
'Dubious economic advice: the "Decision" on Sizewell
B', Energy Policy.
Lave,
L. B.
(ed.) (1982). Quantitative
Risk Assessment in
Regulation, Brookings Institution, Washing-
ton DC.
Layfield,
Sir Frank
(1987).
Sizewell B Public
Enquiry, Report by
Sir Frank
Layfield,
8 vols,
Department
of
Energy, HMSO,
London.
Longhurst, J.
W.
S.,
S. E. Green and D. S. Lee
(1987).
Acid
deposition
in the Northern
Hemisphere,
Acid rain Information
Centre,
Manchester
Polytechnic.
Maler,
K.-G.
(1989).
'The acid rain
game, 2', paper presented
at Conference in
Wageningen,
mimeo,
Stockholm School of Economics.
-
(1990).
'International Environmental
Problems', Oxford
Review
of
Economic
Policy.
Newbery,
D. M.
(1990). 'Optimal
Trade Taxes on
Agriculture
in
Developing Countries',
Economic
Journal Supplement.
Newbery,
D. M. and
J.
E.
Stiglitz (1981).
The
Theory of Commodity
Price
Stabilization, Oxford;
Clarendon Press.
OECD
(1981).
The costs and
benefits of sulphur
dioxide control - a
methodological study, OECD,
Paris.
-
(1987).
Environmental Data
Compendium, OECD,
Paris.
Park,
C. P.
(1987).
Acid Rain; Rhetoric and
Reality, Methuen,
London.
Pearce,
D. and A.
Markandya (1989).
Environmental
Policy benefits: Monetary Valuation, OECD,
Paris.
Salop,
S. C. and D. T. Scheffman
(1983). 'Raising
Rivals'
Costs',
American Economic
Review, Papers
and
Proceedings.
Schechter, M., Kim,
M. and
Golan,
L.
(1989). 'Valuing
a
public good:
direct and indirect valuation
approaches
to the measurement of the benefits of
pollution abatement',
in H. Fomer and E.
van Ierland
(eds.)
Valuation Methods and
Policy Making
in Environmental
Economics, Amsterdam,
Elsevier.
Summers,
R. and A. Heston
(1988).
'A New Set of International
Comparisons
of Real Product
and Prices for 130
Countries, 1950-1985',
The Review
of
Income and Wealth.
UNECE
(1982).
'Effects of
sulphur compounds
on
materials, including
historic and cultural
monuments' Draft
report ENV/IEB/WG1;
Geneva.
-
(1987).
National
Strategies
and
policies for
air
pollution
abatement.
White,
L.
J. (1981). Reforming Regulation:
Processes and
Problems, Prentice-Hall,
Englewood Cliffs,
NJ.
-
(1982).
The
regulation of airpollutant emissionsfrom
motor
vehicles,
American
Enterprise
Institute
for Public
Policy Research, Washington
and London.

Acid Rain, by David Newbery

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Acid Rain, by David Newbery

Caricato da

Copyright:

Formati disponibili

Centre for Economic Policy Research

Center for Economic Studies

Potrebbero piacerti anche